Balasubramaniam Natarajan

Sai Munikoti, Deepesh Agarwal, Laya Das et al.

Deep reinforcement learning (DRL) has empowered a variety of artificial intelligence fields, including pattern recognition, robotics, recommendation-systems, and gaming. Similarly, graph neural networks (GNN) have also demonstrated their superior performance in supervised learning for graph-structured data. In recent times, the fusion of GNN with DRL for graph-structured environments has attracted a lot of attention. This paper provides a comprehensive review of these hybrid works. These works can be classified into two categories: (1) algorithmic enhancement, where DRL and GNN complement each other for better utility; (2) application-specific enhancement, where DRL and GNN support each other. This fusion effectively addresses various complex problems in engineering and life sciences. Based on the review, we further analyze the applicability and benefits of fusing these two domains, especially in terms of increasing generalizability and reducing computational complexity. Finally, the key challenges in integrating DRL and GNN, and potential future research directions are highlighted, which will be of interest to the broader machine learning community.

Sai Munikoti, Balasubramaniam Natarajan, Mahantesh Halappanavar

Influence maximization (IM) is a combinatorial problem of identifying a subset of nodes called the seed nodes in a network (graph), which when activated, provide a maximal spread of influence in the network for a given diffusion model and a budget for seed set size. IM has numerous applications such as viral marketing, epidemic control, sensor placement and other network-related tasks. However, the uses are limited due to the computational complexity of current algorithms. Recently, learning heuristics for IM have been explored to ease the computational burden. However, there are serious limitations in current approaches such as: (1) IM formulations only consider influence via spread and ignore self activation; (2) scalability to large graphs; (3) generalizability across graph families; (4) low computational efficiency with a large running time to identify seed sets for every test network. In this work, we address each of these limitations through a unique approach that involves (1) formulating a generic IM problem as a Markov decision process that handles both intrinsic and influence activations; (2) employing double Q learning to estimate seed nodes; (3) ensuring scalability via sub-graph based representations; and (4) incorporating generalizability via meta-learning across graph families. Extensive experiments are carried out in various standard networks to validate performance of the proposed Graph Meta Reinforcement learning (GraMeR) framework. The results indicate that GraMeR is multiple orders faster and generic than conventional approaches.

Sriram Anbalagan, Deepesh Agarwal, Balasubramaniam Natarajan et al.

A majority of recent advancements related to the fault diagnosis of electrical motors are based on the assumption that training and testing data are drawn from the same distribution. However, the data distribution can vary across different operating conditions during real-world operating scenarios of electrical motors. Consequently, this assumption limits the practical implementation of existing studies for fault diagnosis, as they rely on fully labelled training data spanning all operating conditions and assume a consistent distribution. This is because obtaining a large number of labelled samples for several machines across different fault cases and operating scenarios may be unfeasible. In order to overcome the aforementioned limitations, this work proposes a framework to develop a foundational model for fault diagnosis of electrical motors. It involves building a neural network-based backbone to learn high-level features using self-supervised learning, and then fine-tuning the backbone to achieve specific objectives. The primary advantage of such an approach is that the backbone can be fine-tuned to achieve a wide variety of target tasks using very less amount of training data as compared to traditional supervised learning methodologies. The empirical evaluation demonstrates the effectiveness of the proposed approach by obtaining more than 90\% classification accuracy by fine-tuning the backbone not only across different types of fault scenarios or operating conditions, but also across different machines. This illustrates the promising potential of the proposed approach for cross-machine fault diagnosis tasks in real-world applications.

Sai Munikoti, Deepesh Agarwal, Laya Das et al.

Graph Neural Networks (GNN) provide a powerful framework that elegantly integrates Graph theory with Machine learning for modeling and analysis of networked data. We consider the problem of quantifying the uncertainty in predictions of GNN stemming from modeling errors and measurement uncertainty. We consider aleatoric uncertainty in the form of probabilistic links and noise in feature vector of nodes, while epistemic uncertainty is incorporated via a probability distribution over the model parameters. We propose a unified approach to treat both sources of uncertainty in a Bayesian framework, where Assumed Density Filtering is used to quantify aleatoric uncertainty and Monte Carlo dropout captures uncertainty in model parameters. Finally, the two sources of uncertainty are aggregated to estimate the total uncertainty in predictions of a GNN. Results in the real-world datasets demonstrate that the Bayesian model performs at par with a frequentist model and provides additional information about predictions uncertainty that are sensitive to uncertainties in the data and model.

Sriram Anbalagan, Sai Shashank GP, Deepesh Agarwal et al.

Fault detection and diagnosis of electrical motors are of utmost importance in ensuring the safe and reliable operation of several industrial systems. Detection and diagnosis of faults at the incipient stage allows corrective actions to be taken in order to reduce the severity of faults. The existing data-driven deep learning approaches for machine fault diagnosis rely extensively on huge amounts of labeled samples, where annotations are expensive and time-consuming. However, a major portion of unlabeled condition monitoring data is not exploited in the training process. To overcome this limitation, we propose a foundational model-based Active Learning framework that utilizes less amount of labeled samples, which are most informative and harnesses a large amount of available unlabeled data by effectively combining Active Learning and Contrastive Self-Supervised Learning techniques. It consists of a transformer network-based backbone model trained using an advanced nearest-neighbor contrastive self-supervised learning method. This approach empowers the backbone to learn improved representations of samples derived from raw, unlabeled vibration data. Subsequently, the backbone can undergo fine-tuning to address a range of downstream tasks, both within the same machines and across different machines. The effectiveness of the proposed methodology has been assessed through the fine-tuning of the backbone for multiple target tasks using three distinct machine-bearing fault datasets. The experimental evaluation demonstrates a superior performance as compared to existing state-of-the-art fault diagnosis methods with less amount of labeled data.

Shweta Dahale, Sai Munikoti, Balasubramaniam Natarajan

Uncertainty quantification is a critical yet unsolved challenge for deep learning, especially for the time series imputation with irregularly sampled measurements. To tackle this problem, we propose a novel framework based on the principles of recurrent neural networks and neural stochastic differential equations for reconciling irregularly sampled measurements. We impute measurements at any arbitrary timescale and quantify the uncertainty in the imputations in a principled manner. Specifically, we derive analytical expressions for quantifying and propagating the epistemic and aleatoric uncertainty across time instants. Our experiments on the IEEE 37 bus test distribution system reveal that our framework can outperform state-of-the-art uncertainty quantification approaches for time-series data imputations.

Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Time series forecasting is an important application in various domains such as energy management, traffic planning, financial markets, meteorology, and medicine. However, real-time series data often present intricate temporal variability and sharp fluctuations, which pose significant challenges for time series forecasting. Previous models that rely on 1D time series representations usually struggle with complex temporal variations. To address the limitations of 1D time series, this study introduces the Times2D method that transforms the 1D time series into 2D space. Times2D consists of three main parts: first, a Periodic Decomposition Block (PDB) that captures temporal variations within a period and between the same periods by converting the time series into a 2D tensor in the frequency domain. Second, the First and Second Derivative Heatmaps (FSDH) capture sharp changes and turning points, respectively. Finally, an Aggregation Forecasting Block (AFB) integrates the output tensors from PDB and FSDH for accurate forecasting. This 2D transformation enables the utilization of 2D convolutional operations to effectively capture long and short characteristics of the time series. Comprehensive experimental results across large-scale data in the literature demonstrate that the proposed Times2D model achieves state-of-the-art performance in both short-term and long-term forecasting. The code is available in this repository: https://github.com/Tims2D/Times2D.

Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Residential electricity demand forecasting is critical for efficient energy management and grid stability. Accurate predictions enable utility companies to optimize planning and operations. However, real-world residential electricity demand data often exhibit intricate temporal variability, including multiple seasonalities, periodicities, and abrupt fluctuations, which pose significant challenges for forecasting models. Previous models that rely on statistical methods, recurrent, convolutional neural networks, and transformers often struggle to capture these intricate temporal dynamics. To address these challenges, we propose the Seasonal-Periodic Decomposition Network (SPDNet), a novel deep learning framework consisting of two main modules. The first is the Seasonal-Trend Decomposition Module (STDM), which decomposes the input data into trend, seasonal, and residual components. The second is the Periodical Decomposition Module (PDM), which employs the Fast Fourier Transform to identify the dominant periods. For each dominant period, 1D input data is reshaped into a 2D tensor, where rows represent periods and columns correspond to frequencies. The 2D representations are then processed through three submodules: a 1D convolution to capture sharp fluctuations, a transformer-based encoder to model global patterns, and a 2D convolution to capture interactions between periods. Extensive experiments conducted on real-world residential electricity load data demonstrate that SPDNet outperforms traditional and advanced models in both forecasting accuracy and computational efficiency. The code is available in this repository: https://github.com/Tims2D/SPDNet.

Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Time series forecasting plays a crucial role in many real-world applications, and numerous complex forecasting models have been proposed in recent years. Despite their architectural innovations, most state-of-the-art models report only marginal improvements -- typically just a few thousandths in standard error metrics. These models often incorporate complex data embedding layers to transform raw inputs into higher-dimensional representations to enhance accuracy. But are data embedding techniques actually effective in time series forecasting? Through extensive ablation studies across fifteen state-of-the-art models and four benchmark datasets, we find that removing data embedding layers from many state-of-the-art models does not degrade forecasting performance. In many cases, it improves both accuracy and computational efficiency. The gains from removing embedding layers often exceed the performance differences typically reported between competing models. Code available at: https://github.com/neuripsdataembedidng/DataEmbedding

Priyanka Gautam, Balasubramaniam Natarajan, Sai Munikoti et al.

In an age where information spreads rapidly across social media, effectively identifying influential nodes in dynamic networks is critical. Traditional influence maximization strategies often fail to keep up with rapidly evolving relationships and structures, leading to missed opportunities and inefficiencies. To address this, we propose a novel learning-based approach integrating Graph Neural Networks (GNNs) with Bidirectional Long Short-Term Memory (BiLSTM) models. This hybrid framework captures both structural and temporal dynamics, enabling accurate prediction of candidate nodes for seed set selection. The bidirectional nature of BiLSTM allows our model to analyze patterns from both past and future network states, ensuring adaptability to changes over time. By dynamically adapting to graph evolution at each time snapshot, our approach improves seed set calculation efficiency, achieving an average of 90% accuracy in predicting potential seed nodes across diverse networks. This significantly reduces computational overhead by optimizing the number of nodes evaluated for seed selection. Our method is particularly effective in fields like viral marketing and social network analysis, where understanding temporal dynamics is crucial.

Dylan Wheeler, Balasubramaniam Natarajan

Communication with the goal of accurately conveying meaning, rather than accurately transmitting symbols, has become an area of growing interest. This paradigm, termed semantic communication, typically leverages modern developments in artificial intelligence and machine learning to improve the efficiency and robustness of communication systems. However, a standard model for capturing and quantifying the details of "meaning" is lacking, with many leading approaches to semantic communication adopting a black-box framework with little understanding of what exactly the model is learning. One solution is to utilize the conceptual spaces framework, which models meaning explicitly in a geometric manner. Though prior work studying semantic communication with conceptual spaces has shown promising results, these previous attempts involve hand-crafting a conceptual space model, severely limiting the scalability and practicality of the approach. In this work, we develop a framework for learning a domain of a conceptual space model using only the raw data with high-level property labels. In experiments using the MNIST and CelebA datasets, we show that the domains learned using the framework maintain semantic similarity relations and possess interpretable dimensions.

Deepesh Agarwal, Pravesh Srivastava, Sergio Martin-del-Campo et al.

Active Learning (AL) is a powerful tool to address modern machine learning problems with significantly fewer labeled training instances. However, implementation of traditional AL methodologies in practical scenarios is accompanied by multiple challenges due to the inherent assumptions. There are several hindrances, such as unavailability of labels for the AL algorithm at the beginning; unreliable external source of labels during the querying process; or incompatible mechanisms to evaluate the performance of Active Learner. Inspired by these practical challenges, we present a hybrid query strategy-based AL framework that addresses three practical challenges simultaneously: cold-start, oracle uncertainty and performance evaluation of Active Learner in the absence of ground truth. While a pre-clustering approach is employed to address the cold-start problem, the uncertainty surrounding the expertise of labeler and confidence in the given labels is incorporated to handle oracle uncertainty. The heuristics obtained during the querying process serve as the fundamental premise for accessing the performance of Active Learner. The robustness of the proposed AL framework is evaluated across three different environments and industrial settings. The results demonstrate the capability of the proposed framework to tackle practical challenges during AL implementation in real-world scenarios.

Sai Munikoti, Laya Das, Balasubramaniam Natarajan

In the quest to improve efficiency, interdependence and complexity are becoming defining characteristics of modern complex networks representing engineered and natural systems. Graph theory is a widely used framework for modeling such complex networks and to evaluate their robustness to disruptions. Particularly, identification of critical nodes/links in a graph can facilitate the enhancement of graph (system) robustness and characterize crucial factors of system performance. Most existing methods of critical node identification are based on an iterative approach that explores each node/link of a graph. These methods suffer from high computational complexity and the resulting analysis is network specific. Additionally, uncertainty associated with the underlying graphical model further limits the potential value of these traditional approaches. To overcome these challenges, we propose a Bayesian graph neural network based node classification framework that is computationally efficient and systematically incorporates uncertainties. Instead of utilizing the observed graph for training the model, a MAP estimate of the graph is computed based on the observed topology and node target labels. Further, a Monte-Carlo (MC) dropout algorithm is incorporated to account for the epistemic uncertainty. The fidelity and the gain in computational complexity offered by the Bayesian framework is illustrated using simulation results.

Sai Munikoti, Laya Das, Balasubramaniam Natarajan

Identifying critical nodes and links in graphs is a crucial task. These nodes/links typically represent critical elements/communication links that play a key role in a system's performance. However, a majority of the methods available in the literature on the identification of critical nodes/links are based on an iterative approach that explores each node/link of a graph at a time, repeating for all nodes/links in the graph. Such methods suffer from high computational complexity and the resulting analysis is also network-specific. To overcome these challenges, this article proposes a scalable and generic graph neural network (GNN) based framework for identifying critical nodes/links in large complex networks. The proposed framework defines a GNN based model that learns the node/link criticality score on a small representative subset of nodes/links. An appropriately trained model can be employed to predict the scores of unseen nodes/links in large graphs and consequently identify the most critical ones. The scalability of the framework is demonstrated through prediction of nodes/links scores in large scale synthetic and real-world networks. The proposed approach is fairly accurate in approximating the criticality scores and offers a significant computational advantage over conventional approaches.

Reza Barazideh, Omid Semiari, Solmaz Niknam et al.

Emerging wireless services with extremely high data rate requirements, such as real-time extended reality applications, mandate novel solutions to further increase the capacity of future wireless networks. In this regard, leveraging large available bandwidth at terahertz frequency bands is seen as a key enabler. To overcome the large propagation loss at these very high frequencies, it is inevitable to manage transmissions over highly directional links. However, uncoordinated directional transmissions by a large number of users can cause substantial interference in terahertz networks. While such interference will be received over short random time intervals, the received power can be large. In this work, a new framework based on reinforcement learning is proposed that uses an adaptive multi-thresholding strategy to efficiently detect and mitigate the intermittent interference from directional links in the time domain. To find the optimal thresholds, the problem is formulated as a multidimensional multi-armed bandit system. Then, an algorithm is proposed that allows the receiver to learn the optimal thresholds with very low complexity. Another key advantage of the proposed approach is that it does not rely on any prior knowledge about the interference statistics, and hence, it is suitable for interference mitigation in dynamic scenarios. Simulation results confirm the superior bit-error-rate performance of the proposed method compared with two traditional time-domain interference mitigation approaches.