Balaraman Ravindran

Adithya Ramesh, Balaraman Ravindran

We apply reinforcement learning (RL) to robotics tasks. One of the drawbacks of traditional RL algorithms has been their poor sample efficiency. One approach to improve the sample efficiency is model-based RL. In our model-based RL algorithm, we learn a model of the environment, essentially its transition dynamics and reward function, use it to generate imaginary trajectories and backpropagate through them to update the policy, exploiting the differentiability of the model. Intuitively, learning more accurate models should lead to better model-based RL performance. Recently, there has been growing interest in developing better deep neural network based dynamics models for physical systems, by utilizing the structure of the underlying physics. We focus on robotic systems undergoing rigid body motion without contacts. We compare two versions of our model-based RL algorithm, one which uses a standard deep neural network based dynamics model and the other which uses a much more accurate, physics-informed neural network based dynamics model. We show that, in model-based RL, model accuracy mainly matters in environments that are sensitive to initial conditions, where numerical errors accumulate fast. In these environments, the physics-informed version of our algorithm achieves significantly better average-return and sample efficiency. In environments that are not sensitive to initial conditions, both versions of our algorithm achieve similar average-return, while the physics-informed version achieves better sample efficiency. We also show that, in challenging environments, physics-informed model-based RL achieves better average-return than state-of-the-art model-free RL algorithms such as Soft Actor-Critic, as it computes the policy-gradient analytically, while the latter estimates it through sampling.

Aravind Venugopal, Stephanie Milani, Fei Fang et al.

Multi-agent reinforcement learning (MARL) methods often suffer from high sample complexity, limiting their use in real-world problems where data is sparse or expensive to collect. Although latent-variable world models have been employed to address this issue by generating abundant synthetic data for MARL training, most of these models cannot encode vital global information available during training into their latent states, which hampers learning efficiency. The few exceptions that incorporate global information assume centralized execution of their learned policies, which is impractical in many applications with partial observability. We propose a novel model-based MARL algorithm, MABL (Multi-Agent Bi-Level world model), that learns a bi-level latent-variable world model from high-dimensional inputs. Unlike existing models, MABL is capable of encoding essential global information into the latent states during training while guaranteeing the decentralized execution of learned policies. For each agent, MABL learns a global latent state at the upper level, which is used to inform the learning of an agent latent state at the lower level. During execution, agents exclusively use lower-level latent states and act independently. Crucially, MABL can be combined with any model-free MARL algorithm for policy learning. In our empirical evaluation with complex discrete and continuous multi-agent tasks including SMAC, Flatland, and MAMuJoCo, MABL surpasses SOTA multi-agent latent-variable world models in both sample efficiency and overall performance.

Sahil Girhepuje, Anmol Goel, Gokul S Krishnan et al.

Recent advances and applications of language technology and artificial intelligence have enabled much success across multiple domains like law, medical and mental health. AI-based Language Models, like Judgement Prediction, have recently been proposed for the legal sector. However, these models are strife with encoded social biases picked up from the training data. While bias and fairness have been studied across NLP, most studies primarily locate themselves within a Western context. In this work, we present an initial investigation of fairness from the Indian perspective in the legal domain. We highlight the propagation of learnt algorithmic biases in the bail prediction task for models trained on Hindi legal documents. We evaluate the fairness gap using demographic parity and show that a decision tree model trained for the bail prediction task has an overall fairness disparity of 0.237 between input features associated with Hindus and Muslims. Additionally, we highlight the need for further research and studies in the avenues of fairness/bias in applying AI in the legal sector with a specific focus on the Indian context.

Shreya Goyal, Sumanth Doddapaneni, Mitesh M. Khapra et al.

In the past few years, it has become increasingly evident that deep neural networks are not resilient enough to withstand adversarial perturbations in input data, leaving them vulnerable to attack. Various authors have proposed strong adversarial attacks for computer vision and Natural Language Processing (NLP) tasks. As a response, many defense mechanisms have also been proposed to prevent these networks from failing. The significance of defending neural networks against adversarial attacks lies in ensuring that the model's predictions remain unchanged even if the input data is perturbed. Several methods for adversarial defense in NLP have been proposed, catering to different NLP tasks such as text classification, named entity recognition, and natural language inference. Some of these methods not only defend neural networks against adversarial attacks but also act as a regularization mechanism during training, saving the model from overfitting. This survey aims to review the various methods proposed for adversarial defenses in NLP over the past few years by introducing a novel taxonomy. The survey also highlights the fragility of advanced deep neural networks in NLP and the challenges involved in defending them.

Jatin Chauhan, Aravindan Raghuveer, Rishi Saket et al.

We formulate a new inference task in the domain of multivariate time series forecasting (MTSF), called Variable Subset Forecast (VSF), where only a small subset of the variables is available during inference. Variables are absent during inference because of long-term data loss (eg. sensor failures) or high -> low-resource domain shift between train / test. To the best of our knowledge, robustness of MTSF models in presence of such failures, has not been studied in the literature. Through extensive evaluation, we first show that the performance of state of the art methods degrade significantly in the VSF setting. We propose a non-parametric, wrapper technique that can be applied on top any existing forecast models. Through systematic experiments across 4 datasets and 5 forecast models, we show that our technique is able to recover close to 95\% performance of the models even when only 15\% of the original variables are present.

Kushal Chauhan, Soumya Chatterjee, Akash Reddy et al.

The options framework in Hierarchical Reinforcement Learning breaks down overall goals into a combination of options or simpler tasks and associated policies, allowing for abstraction in the action space. Ideally, these options can be reused across different higher-level goals; indeed, such reuse is necessary to realize the vision of a continual learning agent that can effectively leverage its prior experience. Previous approaches have only proposed limited forms of transfer of prelearned options to new task settings. We propose a novel option indexing approach to hierarchical learning (OI-HRL), where we learn an affinity function between options and the items present in the environment. This allows us to effectively reuse a large library of pretrained options, in zero-shot generalization at test time, by restricting goal-directed learning to only those options relevant to the task at hand. We develop a meta-training loop that learns the representations of options and environments over a series of HRL problems, by incorporating feedback about the relevance of retrieved options to the higher-level goal. We evaluate OI-HRL in two simulated settings - the CraftWorld and AI2THOR environments - and show that we achieve performance competitive with oracular baselines, and substantial gains over a baseline that has the entire option pool available for learning the hierarchical policy.

Saket Gurukar, Shaileshh Bojja Venkatakrishnan, Balaraman Ravindran et al.

Graph convolutional networks (GCNs) have achieved huge success in several machine learning (ML) tasks on graph-structured data. Recently, several sampling techniques have been proposed for the efficient training of GCNs and to improve the performance of GCNs on ML tasks. Specifically, the subgraph-based sampling approaches such as ClusterGCN and GraphSAINT have achieved state-of-the-art performance on the node classification tasks. These subgraph-based sampling approaches rely on heuristics -- such as graph partitioning via edge cuts -- to identify clusters that are then treated as minibatches during GCN training. In this work, we hypothesize that rather than relying on such heuristics, one can learn a reinforcement learning (RL) policy to compute efficient clusters that lead to effective GCN performance. To that end, we propose PolicyClusterGCN, an online RL framework that can identify good clusters for GCN training. We develop a novel Markov Decision Process (MDP) formulation that allows the policy network to predict ``importance" weights on the edges which are then utilized by a clustering algorithm (Graclus) to compute the clusters. We train the policy network using a standard policy gradient algorithm where the rewards are computed from the classification accuracies while training GCN using clusters given by the policy. Experiments on six real-world datasets and several synthetic datasets show that PolicyClusterGCN outperforms existing state-of-the-art models on node classification task.

Richa Verma, Balaraman Ravindran

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a powerful paradigm for reasoning in language models, with GRPO as its primary example. However, GRPO requires continuous online rollout generation, making it computationally expensive and difficult to scale. While Direct Preference Optimization (DPO) offers a stable and efficient offline alternative, it is typically expected to underperform w.r.t. online RL methods such as GRPO when trained on rollouts from a cold supervised fine-tuned (SFT) policy. We introduce G2D (GRPO to DPO)}, a three-stage pipeline that performs a short GRPO warm-up, constructs a static preference dataset, and fine-tunes a model offline with DPO. Across a set of values of the number of online steps (K) in GRPO on Qwen2.5-7B and Llama-3.1-8B, we find that offline DPO with moderate warm-up matches or outperforms GRPO at substantially lower compute cost in our setting. On Qwen2.5-7B, G2D at K=150 achieves 62.4% on MATH-500, outperforming GRPO (51.6%) by 10.8% at ~4x lower compute. On Llama-3.1-8B, G2D at K=500 achieves 49.4%, surpassing GRPO in our experimental setting. We show that performance is not governed by the number of preference pairs, which does not vary much w.r.t. K, but by their informativeness. Moderate warm-up produces rollouts with calibrated uncertainty, yielding stronger contrastive signal, while excessive warm-up leads to overconfident policies and less informative data. Our results recast the offline-online gap in RLVR as primarily a data informativeness problem, and identify short online RL warm-up with appropriate difficulty calibration of the fine-tuning dataset as a compute-efficient alternative to online RL.

Richa Verma, Bavish Kulur, Sanjay Chawla et al.

We address the problem of making a pre-trained reinforcement learning (RL) policy safety-aware by incorporating cost constraints without retraining it from scratch. While costs could be numerically encoded, we assume a more general setting is when costs are provided as preferences. Given a reward-optimized policy and a small dataset of preferred (low-cost) and dispreferred (high-cost) trajectories, our goal is to fine-tune the policy to generate low-cost behaviors while retaining high rewards. Unlike standard RLHF in language models, where preferences are defined over responses to the same prompt, our setting involves trajectory-level preferences in continuous control environments. We introduce PREFINE: Preference-based Implicit Reward and Cost Fine-Tuning for Safety Alignment which is a preference-based fine-tuning method that adapts Direct Preference Optimization (DPO), which is now widely used for LLM fine-tuning, to the sequential decision making setting. PREFINE constructs policy-sampled counterfactual trajectories to establish meaningful preference contrasts and jointly optimizes for reward retention and safety alignment. Empirically, PREFINE reduces constraint violations and catastrophic failures by over 60% while maintaining original reward behavior. PREFINE produces policies that achieve low-cost, high-reward performance with significantly improved data and computational efficiency compared to full offline RL or imitation learning, bridging preference alignment and safe policy adaptation in continuous domains.

Omkar Shelke, Pranavi Pathakota, Anandsingh Chauhan et al.

This paper presents an integrated algorithmic framework for minimising product delivery costs in e-commerce (known as the cost-to-serve or C2S). One of the major challenges in e-commerce is the large volume of spatio-temporally diverse orders from multiple customers, each of which has to be fulfilled from one of several warehouses using a fleet of vehicles. This results in two levels of decision-making: (i) selection of a fulfillment node for each order (including the option of deferral to a future time), and then (ii) routing of vehicles (each of which can carry multiple orders originating from the same warehouse). We propose an approach that combines graph neural networks and reinforcement learning to train the node selection and vehicle routing agents. We include real-world constraints such as warehouse inventory capacity, vehicle characteristics such as travel times, service times, carrying capacity, and customer constraints including time windows for delivery. The complexity of this problem arises from the fact that outcomes (rewards) are driven both by the fulfillment node mapping as well as the routing algorithms, and are spatio-temporally distributed. Our experiments show that this algorithmic pipeline outperforms pure heuristic policies.

Neeraja Kirtane, Jeshuren Chelladurai, Balaraman Ravindran et al.

Node classification is an important task to solve in graph-based learning. Even though a lot of work has been done in this field, imbalance is neglected. Real-world data is not perfect, and is imbalanced in representations most of the times. Apart from text and images, data can be represented using graphs, and thus addressing the imbalance in graphs has become of paramount importance. In the context of node classification, one class has less examples than others. Changing data composition is a popular way to address the imbalance in node classification. This is done by resampling the data to balance the dataset. However, that can sometimes lead to loss of information or add noise to the dataset. Therefore, in this work, we implicitly solve the problem by changing the model loss. Specifically, we study how attention networks can help tackle imbalance. Moreover, we observe that using a regularizer to assign larger weights to minority nodes helps to mitigate this imbalance. We achieve State of the Art results than the existing methods on several standard citation benchmark datasets.

Krithi Shailya, Shreya Rajpal, Gokul S Krishnan et al.

As Large Language Models (LLMs) become increasingly integrated into high-stakes domains, there have been several approaches proposed toward generating natural language explanations. These explanations are crucial for enhancing the interpretability of a model, especially in sensitive domains like healthcare, where transparency and reliability are key. In light of such explanations being generated by LLMs and its known concerns, there is a growing need for robust evaluation frameworks to assess model-generated explanations. Natural Language Generation metrics like BLEU and ROUGE capture syntactic and semantic accuracies but overlook other crucial aspects such as factual accuracy, consistency, and faithfulness. To address this gap, we propose a general framework for quantifying trustworthiness of natural language explanations, balancing Plausibility and Faithfulness, to derive a comprehensive Language Explanation Trustworthiness Score (LExT) (The code and set up to reproduce our experiments are publicly available at https://github.com/cerai-iitm/LExT). Applying our domain-agnostic framework to the healthcare domain using public medical datasets, we evaluate six models, including domain-specific and general-purpose models. Our findings demonstrate significant differences in their ability to generate trustworthy explanations. On comparing these explanations, we make interesting observations such as inconsistencies in Faithfulness demonstrated by general-purpose models and their tendency to outperform domain-specific fine-tuned models. This work further highlights the importance of using a tailored evaluation framework to assess natural language explanations in sensitive fields, providing a foundation for improving the trustworthiness and transparency of language models in healthcare and beyond.

Jeshuren Chelladurai, Sudarsun Santhiappan, Balaraman Ravindran

Automated medical coding is a process of codifying clinical notes to appropriate diagnosis and procedure codes automatically from the standard taxonomies such as ICD (International Classification of Diseases) and CPT (Current Procedure Terminology). The manual coding process involves the identification of entities from the clinical notes followed by querying a commercial or non-commercial medical codes Information Retrieval (IR) system that follows the Centre for Medicare and Medicaid Services (CMS) guidelines. We propose to automate this manual process by automatically constructing a query for the IR system using the entities auto-extracted from the clinical notes. We propose \textbf{GrabQC}, a \textbf{Gra}ph \textbf{b}ased \textbf{Q}uery \textbf{C}ontextualization method that automatically extracts queries from the clinical text, contextualizes the queries using a Graph Neural Network (GNN) model and obtains the ICD Codes using an external IR system. We also propose a method for labelling the dataset for training the model. We perform experiments on two datasets of clinical text in three different setups to assert the effectiveness of our approach. The experimental results show that our proposed method is better than the compared baselines in all three settings.

Yogesh Tripathi, Raghav Donakanti, Sahil Girhepuje et al.

Recent advancements in language technology and Artificial Intelligence have resulted in numerous Language Models being proposed to perform various tasks in the legal domain ranging from predicting judgments to generating summaries. Despite their immense potential, these models have been proven to learn and exhibit societal biases and make unfair predictions. In this study, we explore the ability of Large Language Models (LLMs) to perform legal tasks in the Indian landscape when social factors are involved. We present a novel metric, $β$-weighted $\textit{Legal Safety Score ($LSS_β$)}$, which encapsulates both the fairness and accuracy aspects of the LLM. We assess LLMs' safety by considering its performance in the $\textit{Binary Statutory Reasoning}$ task and its fairness exhibition with respect to various axes of disparities in the Indian society. Task performance and fairness scores of LLaMA and LLaMA--2 models indicate that the proposed $LSS_β$ metric can effectively determine the readiness of a model for safe usage in the legal sector. We also propose finetuning pipelines, utilising specialised legal datasets, as a potential method to mitigate bias and improve model safety. The finetuning procedures on LLaMA and LLaMA--2 models increase the $LSS_β$, improving their usability in the Indian legal domain. Our code is publicly released.

Clifford F, Devika Jay, Abhishek Sarkar et al.

As Reinforcement Learning (RL) agents are increasingly deployed in real-world applications, ensuring their behavior is transparent and trustworthy is paramount. A key component of trust is explainability, yet much of the work in Explainable RL (XRL) focuses on local, single-step decisions. This paper addresses the critical need for explaining an agent's long-term behavior through trajectory-level analysis. We introduce a novel framework that ranks entire trajectories by defining and aggregating a new state-importance metric. This metric combines the classic Q-value difference with a "radical term" that captures the agent's affinity to reach its goal, providing a more nuanced measure of state criticality. We demonstrate that our method successfully identifies optimal trajectories from a heterogeneous collection of agent experiences. Furthermore, by generating counterfactual rollouts from critical states within these trajectories, we show that the agent's chosen path is robustly superior to alternatives, thereby providing a powerful "Why this, and not that?" explanation. Our experiments in standard OpenAI Gym environments validate that our proposed importance metric is more effective at identifying optimal behaviors compared to classic approaches, offering a significant step towards trustworthy autonomous systems.

Jashaswimalya Acharjee, Balaraman Ravindran

We present Unified Latent Dynamics (ULD), a novel reinforcement learning algorithm that unifies the efficiency of model-free methods with the representational strengths of model-based approaches, without incurring planning overhead. By embedding state-action pairs into a latent space in which the true value function is approximately linear, our method supports a single set of hyperparameters across diverse domains -- from continuous control with low-dimensional and pixel inputs to high-dimensional Atari games. We prove that, under mild conditions, the fixed point of our embedding-based temporal-difference updates coincides with that of a corresponding linear model-based value expansion, and we derive explicit error bounds relating embedding fidelity to value approximation quality. In practice, ULD employs synchronized updates of encoder, value, and policy networks, auxiliary losses for short-horizon predictive dynamics, and reward-scale normalization to ensure stable learning under sparse rewards. Evaluated on 80 environments spanning Gym locomotion, DeepMind Control (proprioceptive and visual), and Atari, our approach matches or exceeds the performance of specialized model-free and general model-based baselines -- achieving cross-domain competence with minimal tuning and a fraction of the parameter footprint. These results indicate that value-aligned latent representations alone can deliver the adaptability and sample efficiency traditionally attributed to full model-based planning.

Returaj Burnwal, Nirav Pravinbhai Bhatt, Balaraman Ravindran

This work addresses the problem of offline safe imitation learning (IL), where the goal is to learn safe and reward-maximizing policies from demonstrations that do not have per-timestep safety cost or reward information. In many real-world domains, online learning in the environment can be risky, and specifying accurate safety costs can be difficult. However, it is often feasible to collect trajectories that reflect undesirable or unsafe behavior, implicitly conveying what the agent should avoid. We refer to these as non-preferred trajectories. We propose a novel offline safe IL algorithm, OSIL, that infers safety from non-preferred demonstrations. We formulate safe policy learning as a Constrained Markov Decision Process (CMDP). Instead of relying on explicit safety cost and reward annotations, OSIL reformulates the CMDP problem by deriving a lower bound on reward maximizing objective and learning a cost model that estimates the likelihood of non-preferred behavior. Our approach allows agents to learn safe and reward-maximizing behavior entirely from offline demonstrations. We empirically demonstrate that our approach can learn safer policies that satisfy cost constraints without degrading the reward performance, thus outperforming several baselines.

Returaj Burnwal, Nirav Pravinbhai Bhatt, Balaraman Ravindran

In this work, we study the problem of offline safe imitation learning (IL). In many real-world settings, online interactions can be risky, and accurately specifying the reward and the safety cost information at each timestep can be difficult. However, it is often feasible to collect trajectories reflecting undesirable or risky behavior, implicitly conveying the behavior the agent should avoid. We refer to these trajectories as non-preferred trajectories. Unlike standard IL, which aims to mimic demonstrations, our agent must also learn to avoid risky behavior using non-preferred trajectories. In this paper, we propose a novel approach, SafeMIL, to learn a parameterized cost that predicts if the state-action pair is risky via Multiple Instance Learning. The learned cost is then used to avoid non-preferred behaviors, resulting in a policy that prioritizes safety. We empirically demonstrate that our approach can learn a safer policy that satisfies cost constraints without degrading the reward performance, thereby outperforming several baselines.

Shreyash Adappanavar, Krithi Shailya, Gokul S Krishnan et al.

The deployment of Large Language Models (LLMs) in high-stakes medical settings poses a critical AI alignment challenge, as models can inherit and amplify societal biases, leading to significant disparities. Existing fairness evaluation methods fall short in these contexts as they typically use simplistic metrics that overlook the multi-dimensional nature of medical harms. This also promotes models that are fair only because they are clinically inert, defaulting to safe but potentially inaccurate outputs. To address this gap, our contributions are mainly two-fold: first, we construct two large-scale, controlled benchmarks (ED-Triage and Opioid Analgesic Recommendation) from MIMIC-IV, comprising over 50,000 prompts with twelve race x gender variants and three context tiers. Second, we propose a multi-metric framework - Multi-faceted Fairness Assessment based on hARMs ($mFARM$) to audit fairness for three distinct dimensions of disparity (Allocational, Stability, and Latent) and aggregate them into an $mFARM$ score. We also present an aggregated Fairness-Accuracy Balance (FAB) score to benchmark and observe trade-offs between fairness and prediction accuracy. We empirically evaluate four open-source LLMs (Mistral-7B, BioMistral-7B, Qwen-2.5-7B, Bio-LLaMA3-8B) and their finetuned versions under quantization and context variations. Our findings showcase that the proposed $mFARM$ metrics capture subtle biases more effectively under various settings. We find that most models maintain robust performance in terms of $mFARM$ score across varying levels of quantization but deteriorate significantly when the context is reduced. Our benchmarks and evaluation code are publicly released to enhance research in aligned AI for healthcare.

Krithi Shailya, Akhilesh Kumar Mishra, Gokul S Krishnan et al.

Large Language Models (LLMs) are increasingly used as daily recommendation systems for tasks like education planning, yet their recommendations risk perpetuating societal biases. This paper empirically examines geographic, demographic, and economic biases in university and program suggestions from three open-source LLMs: LLaMA-3.1-8B, Gemma-7B, and Mistral-7B. Using 360 simulated user profiles varying by gender, nationality, and economic status, we analyze over 25,000 recommendations. Results show strong biases: institutions in the Global North are disproportionately favored, recommendations often reinforce gender stereotypes, and institutional repetition is prevalent. While LLaMA-3.1 achieves the highest diversity, recommending 481 unique universities across 58 countries, systemic disparities persist. To quantify these issues, we propose a novel, multi-dimensional evaluation framework that goes beyond accuracy by measuring demographic and geographic representation. Our findings highlight the urgent need for bias consideration in educational LMs to ensure equitable global access to higher education.

Deepak Maurya, Balaraman Ravindran

Link prediction in graphs is studied by modeling the dyadic interactions among two nodes. The relationships can be more complex than simple dyadic interactions and could require the user to model super-dyadic associations among nodes. Such interactions can be modeled using a hypergraph, which is a generalization of a graph where a hyperedge can connect more than two nodes. In this work, we consider the problem of hyperedge prediction in a $k-$uniform hypergraph. We utilize the tensor-based representation of hypergraphs and propose a novel interpretation of the tensor eigenvectors. This is further used to propose a hyperedge prediction algorithm. The proposed algorithm utilizes the \textit{Fiedler} eigenvector computed using tensor eigenvalue decomposition of hypergraph Laplacian. The \textit{Fiedler} eigenvector is used to evaluate the construction cost of new hyperedges, which is further utilized to determine the most probable hyperedges to be constructed. The functioning and efficacy of the proposed method are illustrated using some example hypergraphs and a few real datasets. The code for the proposed method is available on https://github.com/d-maurya/hypred_ tensorEVD

Anirban Santara, Sohan Rudra, Sree Aditya Buridi et al.

In this work, we present MADRaS, an open-source multi-agent driving simulator for use in the design and evaluation of motion planning algorithms for autonomous driving. MADRaS provides a platform for constructing a wide variety of highway and track driving scenarios where multiple driving agents can train for motion planning tasks using reinforcement learning and other machine learning algorithms. MADRaS is built on TORCS, an open-source car-racing simulator. TORCS offers a variety of cars with different dynamic properties and driving tracks with different geometries and surface properties. MADRaS inherits these functionalities from TORCS and introduces support for multi-agent training, inter-vehicular communication, noisy observations, stochastic actions, and custom traffic cars whose behaviours can be programmed to simulate challenging traffic conditions encountered in the real world. MADRaS can be used to create driving tasks whose complexities can be tuned along eight axes in well-defined steps. This makes it particularly suited for curriculum and continual learning. MADRaS is lightweight and it provides a convenient OpenAI Gym interface for independent control of each car. Apart from the primitive steering-acceleration-brake control mode of TORCS, MADRaS offers a hierarchical track-position -- speed control that can potentially be used to achieve better generalization. MADRaS uses multiprocessing to run each agent as a parallel process for efficiency and integrates well with popular reinforcement learning libraries like RLLib.

Akash Kumar Mohankumar, Preksha Nema, Sharan Narasimhan et al.

Recent studies on interpretability of attention distributions have led to notions of faithful and plausible explanations for a model's predictions. Attention distributions can be considered a faithful explanation if a higher attention weight implies a greater impact on the model's prediction. They can be considered a plausible explanation if they provide a human-understandable justification for the model's predictions. In this work, we first explain why current attention mechanisms in LSTM based encoders can neither provide a faithful nor a plausible explanation of the model's predictions. We observe that in LSTM based encoders the hidden representations at different time-steps are very similar to each other (high conicity) and attention weights in these situations do not carry much meaning because even a random permutation of the attention weights does not affect the model's predictions. Based on experiments on a wide variety of tasks and datasets, we observe attention distributions often attribute the model's predictions to unimportant words such as punctuation and fail to offer a plausible explanation for the predictions. To make attention mechanisms more faithful and plausible, we propose a modified LSTM cell with a diversity-driven training objective that ensures that the hidden representations learned at different time steps are diverse. We show that the resulting attention distributions offer more transparency as they (i) provide a more precise importance ranking of the hidden states (ii) are better indicative of words important for the model's predictions (iii) correlate better with gradient-based attribution methods. Human evaluations indicate that the attention distributions learned by our model offer a plausible explanation of the model's predictions. Our code has been made publicly available at https://github.com/akashkm99/Interpretable-Attention

Preksha Nema, Akash Kumar Mohankumar, Mitesh M. Khapra et al.

In this work, we focus on the task of Automatic Question Generation (AQG) where given a passage and an answer the task is to generate the corresponding question. It is desired that the generated question should be (i) grammatically correct (ii) answerable from the passage and (iii) specific to the given answer. An analysis of existing AQG models shows that they produce questions which do not adhere to one or more of {the above-mentioned qualities}. In particular, the generated questions look like an incomplete draft of the desired question with a clear scope for refinement. {To alleviate this shortcoming}, we propose a method which tries to mimic the human process of generating questions by first creating an initial draft and then refining it. More specifically, we propose Refine Network (RefNet) which contains two decoders. The second decoder uses a dual attention network which pays attention to both (i) the original passage and (ii) the question (initial draft) generated by the first decoder. In effect, it refines the question generated by the first decoder, thereby making it more correct and complete. We evaluate RefNet on three datasets, \textit{viz.}, SQuAD, HOTPOT-QA, and DROP, and show that it outperforms existing state-of-the-art methods by 7-16\% on all of these datasets. Lastly, we show that we can improve the quality of the second decoder on specific metrics, such as, fluency and answerability by explicitly rewarding revisions that improve on the corresponding metric during training. The code has been made publicly available \footnote{https://github.com/PrekshaNema25/RefNet-QG}

Yoshua Bengio, Sören Mindermann, Daniel Privitera et al. · eth-zurich, mit

The first International AI Safety Report comprehensively synthesizes the current evidence on the capabilities, risks, and safety of advanced AI systems. The report was mandated by the nations attending the AI Safety Summit in Bletchley, UK. Thirty nations, the UN, the OECD, and the EU each nominated a representative to the report's Expert Advisory Panel. A total of 100 AI experts contributed, representing diverse perspectives and disciplines. Led by the report's Chair, these independent experts collectively had full discretion over the report's content.

Atharvan Dogra, Krishna Pillutla, Ameet Deshpande et al.

We explore the ability of large language models (LLMs) to engage in subtle deception through strategically phrasing and intentionally manipulating information. This harmful behavior can be hard to detect, unlike blatant lying or unintentional hallucination. We build a simple testbed mimicking a legislative environment where a corporate \textit{lobbyist} module is proposing amendments to bills that benefit a specific company while evading identification of this benefactor. We use real-world legislative bills matched with potentially affected companies to ground these interactions. Our results show that LLM lobbyists can draft subtle phrasing to avoid such identification by strong LLM-based detectors. Further optimization of the phrasing using LLM-based re-planning and re-sampling increases deception rates by up to 40 percentage points. Our human evaluations to verify the quality of deceptive generations and their retention of self-serving intent show significant coherence with our automated metrics and also help in identifying certain strategies of deceptive phrasing. This study highlights the risk of LLMs' capabilities for strategic phrasing through seemingly neutral language to attain self-serving goals. This calls for future research to uncover and protect against such subtle deception.

Gokul S Krishnan, Sarala Padi, Craig S. Greenberg et al.

Emotion Recognition in Conversations (ERC) is a critical aspect of affective computing, and it has many practical applications in healthcare, education, chatbots, and social media platforms. Earlier approaches for ERC analysis involved modeling both speaker and long-term contextual information using graph neural network architectures. However, it is ideal to deploy speaker-independent models for real-world applications. Additionally, long context windows can potentially create confusion in recognizing the emotion of an utterance in a conversation. To overcome these limitations, we propose novel line conversation graph convolutional network (LineConGCN) and graph attention (LineConGAT) models for ERC analysis. These models are speaker-independent and built using a graph construction strategy for conversations -- line conversation graphs (LineConGraphs). The conversational context in LineConGraphs is short-term -- limited to one previous and future utterance, and speaker information is not part of the graph. We evaluate the performance of our proposed models on two benchmark datasets, IEMOCAP and MELD, and show that our LineConGAT model outperforms the state-of-the-art methods with an F1-score of 64.58% and 76.50%. Moreover, we demonstrate that embedding sentiment shift information into line conversation graphs further enhances the ERC performance in the case of GCN models.

Ambreesh Parthasarathy, Chandrasekar Subramanian, Ganesh Senrayan et al.

Restless Multi-Armed Bandits (RMABs) have been successfully applied to resource allocation problems in a variety of settings, including public health. With the rapid development of powerful large language models (LLMs), they are increasingly used to design reward functions to better match human preferences. Recent work has shown that LLMs can be used to tailor automated allocation decisions to community needs using language prompts. However, this has been studied primarily for English prompts and with a focus on task performance only. This can be an issue since grassroots workers, especially in developing countries like India, prefer to work in local languages, some of which are low-resource. Further, given the nature of the problem, biases along population groups unintended by the user are also undesirable. In this work, we study the effects on both task performance and fairness when the DLM algorithm, a recent work on using LLMs to design reward functions for RMABs, is prompted with non-English language commands. Specifically, we run the model on a synthetic environment for various prompts translated into multiple languages. The prompts themselves vary in complexity. Our results show that the LLM-proposed reward functions are significantly better when prompted in English compared to other languages. We also find that the exact phrasing of the prompt impacts task performance. Further, as prompt complexity increases, performance worsens for all languages; however, it is more robust with English prompts than with lower-resource languages. On the fairness side, we find that low-resource languages and more complex prompts are both highly likely to create unfairness along unintended dimensions.

Santhosh G S, Saurav Prakash, Balaraman Ravindran

Large Language Models (LLMs) face a significant bottleneck during autoregressive inference due to the massive memory footprint of the Key-Value (KV) cache. Existing compression techniques like token eviction, quantization, or other low-rank methods often risk information loss, have fixed limits, or introduce significant computational overhead from explicit decompression steps. In this work, we introduce SWAN, a novel, fine-tuning-free framework that eliminates this overhead. Our method uses an offline orthogonal matrix to rotate and prune the KV-cache, which is then used directly in the attention computation without any reconstruction. Our extensive experiments demonstrate that SWAN, augmented with a small dense buffer, offers a robust trade-off, maintaining performance close to the uncompressed baseline even at aggressive 50-60% memory savings per-token on KV-cache. A key advantage is its runtime-tunable compression level, allowing operators to dynamically adjust the memory footprint, a flexibility absent in methods requiring fixed offline configurations. This combination of a decompression-free design, high performance under compression, and adaptability makes SWAN a practical and efficient solution for serving LLMs with long contexts.

Santhosh G S, Akshay Govind S, Gokul S Krishnan et al.

Large Language Models (LLMs) have gained significant traction across critical domains owing to their impressive contextual understanding and generative capabilities. However, their increasing deployment in high stakes applications necessitates rigorous evaluation of embedded biases, particularly in culturally diverse contexts like India where existing embedding-based bias assessment methods often fall short in capturing nuanced stereotypes. We propose an evaluation framework based on a encoder trained using contrastive learning that captures fine-grained bias through embedding similarity. We also introduce a novel dataset - IndiCASA (IndiBias-based Contextually Aligned Stereotypes and Anti-stereotypes) comprising 2,575 human-validated sentences spanning five demographic axes: caste, gender, religion, disability, and socioeconomic status. Our evaluation of multiple open-weight LLMs reveals that all models exhibit some degree of stereotypical bias, with disability related biases being notably persistent, and religion bias generally lower likely due to global debiasing efforts demonstrating the need for fairer model development.

Returaj Burnwal, Hriday Mehta, Nirav Pravinbhai Bhatt et al.

Imitation Learning (IL) algorithms offer an efficient way to train an agent by mimicking an expert's behavior without requiring a reward function. IL algorithms often necessitate access to state and action information from expert demonstrations. Although expert actions can provide detailed guidance, requiring such action information may prove impractical for real-world applications where expert actions are difficult to obtain. To address this limitation, the concept of learning from observation (LfO) or state-only imitation learning (SOIL) has recently gained attention, wherein the imitator only has access to expert state visitation information. In this paper, we present a framework for LfO and use it to survey and classify existing LfO methods in terms of their trajectory construction, assumptions and algorithm's design choices. This survey also draws connections between several related fields like offline RL, model-based RL and hierarchical RL. Finally, we use our framework to identify open problems and suggest future research directions.

Santhosh G S, Saurav Prakash, Balaraman Ravindran

The quadratic complexity of the attention mechanism remains a fundamental barrier to scaling Large Language Models (LLMs) to longer contexts, creating a critical bottleneck in both computation and memory. To address this, we introduce AQUA (Attention via QUery mAgnitudes) a novel and versatile approximation strategy that significantly reduces the cost of attention with a graceful performance trade-off. Our method operates in two phases: an efficient offline step where we compute a universal, language agnostic projection matrix via SVD on a calibration dataset, and an online inference step where we project query and key vectors and dynamically select a sparse subset of dimensions based on the query's magnitude. We provide a formal theoretical analysis of AQUA, establishing the break-even point at which it becomes more computationally efficient than standard attention. Our empirical evaluations on state-of-the-art models like Llama-3.1-8B demonstrate that a 25% reduction in the attention dot-product computation can be achieved with a statistically insignificant impact on performance across a wide range of benchmarks. We further showcase the versatility of AQUA by demonstrating its ability to synergistically accelerate existing token eviction methods like H2O and to directly reduce KV-cache memory size. By offering a controllable knob to balance efficiency and accuracy, AQUA provides a practical and powerful tool for making large-scale LLM inference more accessible and sustainable.

Abhishek Verma, Nallarasan V, Balaraman Ravindran

Transfer learning in Reinforcement Learning (RL) enables agents to leverage knowledge from source tasks to accelerate learning in target tasks. While prior work, such as the Attend, Adapt, and Transfer (A2T) framework, addresses negative transfer and selective transfer, other critical challenges remain underexplored. This paper introduces the Generalized Adaptive Transfer Network (GATN), a deep RL architecture designed to tackle task generalization across domains, robustness to environmental changes, and computational efficiency in transfer. GATN employs a domain-agnostic representation module, a robustness-aware policy adapter, and an efficient transfer scheduler to achieve these goals. We evaluate GATN on diverse benchmarks, including Atari 2600, MuJoCo, and a custom chatbot dialogue environment, demonstrating superior performance in cross-domain generalization, resilience to dynamic environments, and reduced computational overhead compared to baselines. Our findings suggest GATN is a versatile framework for real-world RL applications, such as adaptive chatbots and robotic control.

Abhishek Verma, Nallarasan V, Balaraman Ravindran

Deep Reinforcement Learning (DRL) has achieved remarkable success in complex sequential decision-making tasks, such as playing Atari 2600 games and mastering board games. A critical yet underexplored aspect of DRL is the temporal scale of action execution. We propose a novel paradigm that integrates contextual bandits with DRL to adaptively select action durations, enhancing policy flexibility and computational efficiency. Our approach augments a Deep Q-Network (DQN) with a contextual bandit module that learns to choose optimal action repetition rates based on state contexts. Experiments on Atari 2600 games demonstrate significant performance improvements over static duration baselines, highlighting the efficacy of adaptive temporal abstractions in DRL. This paradigm offers a scalable solution for real-time applications like gaming and robotics, where dynamic action durations are critical.

Sai Kiran Narayanaswami, Gopalakrishnan Srinivasan, Balaraman Ravindran

As machine learning gets deployed more and more widely, and model sizes continue to grow, improving computational efficiency during model inference has become a key challenge. In many commonly used model architectures, including Transformers, a significant portion of the inference computation is comprised of exponential non-linearities such as Softmax. In this work, we develop QuAKE, a collection of novel operators that leverage certain properties of IEEE-754 floating point representations to quickly approximate the exponential function without requiring specialized hardware, extra memory, or precomputation. We propose optimizations that enhance the efficiency of QuAKE in commonly used exponential non-linearities such as Softmax, GELU, and the Logistic function. Our benchmarks demonstrate substantial inference speed improvements between 10% and 35% on server CPUs, and 5% and 45% on embedded and mobile-scale CPUs for a variety of model architectures and sizes. Evaluations of model performance on standard datasets and tasks from various domains show that QuAKE operators are able to provide sizable speed benefits with little to no loss of performance on downstream tasks.

Returaj Burnwal, Anirban Santara, Nirav P. Bhatt et al.

Model predictive control (MPC) is a popular approach for trajectory optimization in practical robotics applications. MPC policies can optimize trajectory parameters under kinodynamic and safety constraints and provide guarantees on safety, optimality, generalizability, interpretability, and explainability. However, some behaviors are complex and it is difficult to hand-craft an MPC objective function. A special class of MPC policies called Learnable-MPC addresses this difficulty using imitation learning from expert demonstrations. However, they require the demonstrator and the imitator agents to be identical which is hard to satisfy in many real world applications of robotics. In this paper, we address the practical problem of training Learnable-MPC policies when the demonstrator and the imitator do not share the same dynamics and their state spaces may have a partial overlap. We propose a novel approach that uses a generative adversarial network (GAN) to minimize the Jensen-Shannon divergence between the state-trajectory distributions of the demonstrator and the imitator. We evaluate our approach on a variety of simulated robotics tasks of DeepMind Control suite and demonstrate the efficacy of our approach at learning the demonstrator's behavior without having to copy their actions.

Sudarsun Santhiappan, Nitin Shravan, Balaraman Ravindran

Classification model selection is a process of identifying a suitable model class for a given classification task on a dataset. Traditionally, model selection is based on cross-validation, meta-learning, and user preferences, which are often time-consuming and resource-intensive. The performance of any machine learning classification task depends on the choice of the model class, the learning algorithm, and the dataset's characteristics. Our work proposes a novel method for automatic classification model selection from a set of candidate model classes by determining the empirical model-fitness for a dataset based only on its clustering indices. Clustering Indices measure the ability of a clustering algorithm to induce good quality neighborhoods with similar data characteristics. We propose a regression task for a given model class, where the clustering indices of a given dataset form the features and the dependent variable represents the expected classification performance. We compute the dataset clustering indices and directly predict the expected classification performance using the learned regressor for each candidate model class to recommend a suitable model class for dataset classification. We evaluate our model selection method through cross-validation with 60 publicly available binary class datasets and show that our top3 model recommendation is accurate for over 45 of 60 datasets. We also propose an end-to-end Automated ML system for data classification based on our model selection method. We evaluate our end-to-end system against popular commercial and noncommercial Automated ML systems using a different collection of 25 public domain binary class datasets. We show that the proposed system outperforms other methods with an excellent average rank of 1.68.

Pavan Ravishankar, Pranshu Malviya, Balaraman Ravindran

In budget-constrained settings aimed at mitigating unfairness, like law enforcement, it is essential to prioritize the sources of unfairness before taking measures to mitigate them in the real world. Unlike previous works, which only serve as a caution against possible discrimination and de-bias data after data generation, this work provides a toolkit to mitigate unfairness during data generation, given by the Unfair Edge Prioritization algorithm, in addition to de-biasing data after generation, given by the Discrimination Removal algorithm. We assume that a non-parametric Markovian causal model representative of the data generation procedure is given. The edges emanating from the sensitive nodes in the causal graph, such as race, are assumed to be the sources of unfairness. We first quantify Edge Flow in any edge X -> Y, which is the belief of observing a specific value of Y due to the influence of a specific value of X along X -> Y. We then quantify Edge Unfairness by formulating a non-parametric model in terms of edge flows. We then prove that cumulative unfairness towards sensitive groups in a decision, like race in a bail decision, is non-existent when edge unfairness is absent. We prove this result for the non-trivial non-parametric model setting when the cumulative unfairness cannot be expressed in terms of edge unfairness. We then measure the Potential to mitigate the Cumulative Unfairness when edge unfairness is decreased. Based on these measurements, we propose the Unfair Edge Prioritization algorithm that can then be used by policymakers. We also propose the Discrimination Removal Procedure that de-biases a data distribution by eliminating optimization constraints that grow exponentially in the number of sensitive attributes and values taken by them. Extensive experiments validate the theorem and specifications used for quantifying the above measures.

Sapana Chaudhary, Balaraman Ravindran

Imitation learning (IL) is a popular approach in the continuous control setting as among other reasons it circumvents the problems of reward mis-specification and exploration in reinforcement learning (RL). In IL from demonstrations, an important challenge is to obtain agent policies that are smooth with respect to the inputs. Learning through imitation a policy that is smooth as a function of a large state-action ($s$-$a$) space (typical of high dimensional continuous control environments) can be challenging. We take a first step towards tackling this issue by using smoothness inducing regularizers on \textit{both} the policy and the cost models of adversarial imitation learning. Our regularizers work by ensuring that the cost function changes in a controlled manner as a function of $s$-$a$ space; and the agent policy is well behaved with respect to the state space. We call our new smooth IL algorithm \textit{Smooth Policy and Cost Imitation Learning} (SPaCIL, pronounced 'Special'). We introduce a novel metric to quantify the smoothness of the learned policies. We demonstrate SPaCIL's superior performance on continuous control tasks from MuJoCo. The algorithm not just outperforms the state-of-the-art IL algorithm on our proposed smoothness metric, but, enjoys added benefits of faster learning and substantially higher average return.

Harsha Kokel, Arjun Manoharan, Sriraam Natarajan et al.

State abstraction enables sample-efficient learning and better task transfer in complex reinforcement learning environments. Recently, we proposed RePReL (Kokel et al. 2021), a hierarchical framework that leverages a relational planner to provide useful state abstractions for learning. We present a brief overview of this framework and the use of a dynamic probabilistic logic model to design these state abstractions. Our experiments show that RePReL not only achieves better performance and efficient learning on the task at hand but also demonstrates better generalization to unseen tasks.

Amrit Diggavi Seshadri, Balaraman Ravindran

Synthesizing high-quality, realistic images from text-descriptions is a challenging task, and current methods synthesize images from text in a multi-stage manner, typically by first generating a rough initial image and then refining image details at subsequent stages. However, existing methods that follow this paradigm suffer from three important limitations. Firstly, they synthesize initial images without attempting to separate image attributes at a word-level. As a result, object attributes of initial images (that provide a basis for subsequent refinement) are inherently entangled and ambiguous in nature. Secondly, by using common text-representations for all regions, current methods prevent us from interpreting text in fundamentally different ways at different parts of an image. Different image regions are therefore only allowed to assimilate the same type of information from text at each refinement stage. Finally, current methods generate refinement features only once at each refinement stage and attempt to address all image aspects in a single shot. This single-shot refinement limits the precision with which each refinement stage can learn to improve the prior image. Our proposed method introduces three novel components to address these shortcomings: (1) An initial generation stage that explicitly generates separate sets of image features for each word n-gram. (2) A spatial dynamic memory module for refinement of images. (3) An iterative multi-headed mechanism to make it easier to improve upon multiple image aspects. Experimental results demonstrate that our Multi-Headed Spatial Dynamic Memory image refinement with our Multi-Tailed Word-level Initial Generation (MSMT-GAN) performs favourably against the previous state of the art on the CUB and COCO datasets.

Anasua Mitra, Priyesh Vijayan, Ranbir Sanasam et al.

Multiplex networks are complex graph structures in which a set of entities are connected to each other via multiple types of relations, each relation representing a distinct layer. Such graphs are used to investigate many complex biological, social, and technological systems. In this work, we present a novel semi-supervised approach for structure-aware representation learning on multiplex networks. Our approach relies on maximizing the mutual information between local node-wise patch representations and label correlated structure-aware global graph representations to model the nodes and cluster structures jointly. Specifically, it leverages a novel cluster-aware, node-contextualized global graph summary generation strategy for effective joint-modeling of node and cluster representations across the layers of a multiplex network. Empirically, we demonstrate that the proposed architecture outperforms state-of-the-art methods in a range of tasks: classification, clustering, visualization, and similarity search on seven real-world multiplex networks for various experiment settings.

Pranshu Malviya, Balaraman Ravindran, Sarath Chandar

When an agent encounters a continual stream of new tasks in the lifelong learning setting, it leverages the knowledge it gained from the earlier tasks to help learn the new tasks better. In such a scenario, identifying an efficient knowledge representation becomes a challenging problem. Most research works propose to either store a subset of examples from the past tasks in a replay buffer, dedicate a separate set of parameters to each task or penalize excessive updates over parameters by introducing a regularization term. While existing methods employ the general task-agnostic stochastic gradient descent update rule, we propose a task-aware optimizer that adapts the learning rate based on the relatedness among tasks. We utilize the directions taken by the parameters during the updates by accumulating the gradients specific to each task. These task-based accumulated gradients act as a knowledge base that is maintained and updated throughout the stream. We empirically show that our proposed adaptive learning rate not only accounts for catastrophic forgetting but also allows positive backward transfer. We also show that our method performs better than several state-of-the-art methods in lifelong learning on complex datasets with a large number of tasks.

Siddharth Nishtala, Lovish Madaan, Aditya Mate et al.

India has a maternal mortality ratio of 113 and child mortality ratio of 2830 per 100,000 live births. Lack of access to preventive care information is a major contributing factor for these deaths, especially in low resource households. We partner with ARMMAN, a non-profit based in India employing a call-based information program to disseminate health-related information to pregnant women and women with recent child deliveries. We analyze call records of over 300,000 women registered in the program created by ARMMAN and try to identify women who might not engage with these call programs that are proven to result in positive health outcomes. We built machine learning based models to predict the long term engagement pattern from call logs and beneficiaries' demographic information, and discuss the applicability of this method in the real world through a pilot validation. Through a pilot service quality improvement study, we show that using our model's predictions to make interventions boosts engagement metrics by 61.37%. We then formulate the intervention planning problem as restless multi-armed bandits (RMABs), and present preliminary results using this approach.

Nahas Pareekutty, Francis James, Balaraman Ravindran et al.

This paper presents a sampling-based method for optimal motion planning in non-holonomic systems in the absence of known cost functions. It uses the principle of learning through experience to deduce the cost-to-go of regions within the workspace. This cost information is used to bias an incremental graph-based search algorithm that produces solution trajectories. Iterative improvement of cost information and search biasing produces solutions that are proven to be asymptotically optimal. The proposed framework builds on incremental Rapidly-exploring Random Trees (RRT) for random sampling-based search and Reinforcement Learning (RL) to learn workspace costs. A series of experiments were performed to evaluate and demonstrate the performance of the proposed method.

Jahnvi Patel, Devika Jay, Balaraman Ravindran et al.

In real time electricity markets, the objective of generation companies while bidding is to maximize their profit. The strategies for learning optimal bidding have been formulated through game theoretical approaches and stochastic optimization problems. Similar studies in reactive power markets have not been reported so far because the network voltage operating conditions have an increased impact on reactive power markets than on active power markets. Contrary to active power markets, the bids of rivals are not directly related to fuel costs in reactive power markets. Hence, the assumption of a suitable probability distribution function is unrealistic, making the strategies adopted in active power markets unsuitable for learning optimal bids in reactive power market mechanisms. Therefore, a bidding strategy is to be learnt from market observations and experience in imperfect oligopolistic competition-based markets. In this paper, a pioneer work on learning optimal bidding strategies from observation and experience in a three-stage reactive power market is reported.

Ashutosh Kakadiya, Sriraam Natarajan, Balaraman Ravindran

Contextual bandits algorithms have become essential in real-world user interaction problems in recent years. However, these algorithms rely on context as attribute value representation, which makes them unfeasible for real-world domains like social networks are inherently relational. We propose Relational Boosted Bandits(RB2), acontextual bandits algorithm for relational domains based on (relational) boosted trees. RB2 enables us to learn interpretable and explainable models due to the more descriptive nature of the relational representation. We empirically demonstrate the effectiveness and interpretability of RB2 on tasks such as link prediction, relational classification, and recommendations.

Deepak Maurya, Balaraman Ravindran

Hypergraphs have gained increasing attention in the machine learning community lately due to their superiority over graphs in capturing super-dyadic interactions among entities. In this work, we propose a novel approach for the partitioning of k-uniform hypergraphs. Most of the existing methods work by reducing the hypergraph to a graph followed by applying standard graph partitioning algorithms. The reduction step restricts the algorithms to capturing only some weighted pairwise interactions and hence loses essential information about the original hypergraph. We overcome this issue by utilizing the tensor-based representation of hypergraphs, which enables us to capture actual super-dyadic interactions. We prove that the hypergraph to graph reduction is a special case of tensor contraction. We extend the notion of minimum ratio-cut and normalized-cut from graphs to hypergraphs and show the relaxed optimization problem is equivalent to tensor eigenvalue decomposition. This novel formulation also enables us to capture different ways of cutting a hyperedge, unlike the existing reduction approaches. We propose a hypergraph partitioning algorithm inspired from spectral graph theory that can accommodate this notion of hyperedge cuts. We also derive a tighter upper bound on the minimum positive eigenvalue of even-order hypergraph Laplacian tensor in terms of its conductance, which is utilized in the partitioning algorithm to approximate the normalized cut. The efficacy of the proposed method is demonstrated numerically on simple hypergraphs. We also show improvement for the min-cut solution on 2-uniform hypergraphs (graphs) over the standard spectral partitioning algorithm.

Anand A. Rajasekar, Karthik Raman, Balaraman Ravindran

One challenging and essential task in biochemistry is the generation of novel molecules with desired properties. Novel molecule generation remains a challenge since the molecule space is difficult to navigate through, and the generated molecules should obey the rules of chemical valency. Through this work, we propose a novel method, which we call unitMCTS, to perform molecule generation by making a unit change to the molecule at every step using Monte Carlo Tree Search. We show that this method outperforms the recently published techniques on benchmark molecular optimization tasks such as QED and penalized logP. We also demonstrate the usefulness of this method in improving molecule properties while being similar to the starting molecule. Given that there is no learning involved, our method finds desired molecules within a shorter amount of time.

Siddharth Nayak, Balaraman Ravindran

The performance of a trained object detection neural network depends a lot on the image quality. Generally, images are pre-processed before feeding them into the neural network and domain knowledge about the image dataset is used to choose the pre-processing techniques. In this paper, we introduce an algorithm called ObjectRL to choose the amount of a particular pre-processing to be applied to improve the object detection performances of pre-trained networks. The main motivation for ObjectRL is that an image which looks good to a human eye may not necessarily be the optimal one for a pre-trained object detector to detect objects.