Boi Faltings

Debjit Paul, Mete Ismayilzada, Maxime Peyrard et al.

Language models (LMs) have recently shown remarkable performance on reasoning tasks by explicitly generating intermediate inferences, e.g., chain-of-thought prompting. However, these intermediate inference steps may be inappropriate deductions from the initial context and lead to incorrect final predictions. Here we introduce REFINER, a framework for finetuning LMs to explicitly generate intermediate reasoning steps while interacting with a critic model that provides automated feedback on the reasoning. Specifically, the critic provides structured feedback that the reasoning LM uses to iteratively improve its intermediate arguments. Empirical evaluations of REFINER on three diverse reasoning tasks show significant improvements over baseline LMs of comparable scale. Furthermore, when using GPT-3.5 or ChatGPT as the reasoner, the trained critic significantly improves reasoning without finetuning the reasoner. Finally, our critic model is trained without expensive human-in-the-loop data but can be substituted with humans at inference time.

Zhuan Shi, Jing Yan, Xiaoli Tang et al.

The increasing sophistication of text-to-image generative models has led to complex challenges in defining and enforcing copyright infringement criteria and protection. Existing methods, such as watermarking and dataset deduplication, fail to provide comprehensive solutions due to the lack of standardized metrics and the inherent complexity of addressing copyright infringement in diffusion models. To deal with these challenges, we propose a Reinforcement Learning-based Copyright Protection(RLCP) method for Text-to-Image Diffusion Model, which minimizes the generation of copyright-infringing content while maintaining the quality of the model-generated dataset. Our approach begins with the introduction of a novel copyright metric grounded in copyright law and court precedents on infringement. We then utilize the Denoising Diffusion Policy Optimization (DDPO) framework to guide the model through a multi-step decision-making process, optimizing it using a reward function that incorporates our proposed copyright metric. Additionally, we employ KL divergence as a regularization term to mitigate some failure modes and stabilize RL fine-tuning. Experiments conducted on 3 mixed datasets of copyright and non-copyright images demonstrate that our approach significantly reduces copyright infringement risk while maintaining image quality.

Diego Antognini, Boi Faltings

Today, recommender systems are an inevitable part of everyone's daily digital routine and are present on most internet platforms. State-of-the-art deep learning-based models require a large number of data to achieve their best performance. Many datasets fulfilling this criterion have been proposed for multiple domains, such as Amazon products, restaurants, or beers. However, works and datasets in the hotel domain are limited: the largest hotel review dataset is below the million samples. Additionally, the hotel domain suffers from a higher data sparsity than traditional recommendation datasets and therefore, traditional collaborative-filtering approaches cannot be applied to such data. In this paper, we propose HotelRec, a very large-scale hotel recommendation dataset, based on TripAdvisor, containing 50 million reviews. To the best of our knowledge, HotelRec is the largest publicly available dataset in the hotel domain (50M versus 0.9M) and additionally, the largest recommendation dataset in a single domain and with textual reviews (50M versus 22M). We release HotelRec for further research: https://github.com/Diego999/HotelRec.

Zeki Doruk Erden, Boi Faltings

This article critically examines the foundational principles of contemporary AI methods, exploring the limitations that hinder its potential. We draw parallels between the modern AI landscape and the 20th-century Modern Synthesis in evolutionary biology, and highlight how advancements in evolutionary theory that augmented the Modern Synthesis, particularly those of Evolutionary Developmental Biology, offer insights that can inform a new design paradigm for AI. By synthesizing findings across AI and evolutionary theory, we propose a pathway to overcome existing limitations, enabling AI to achieve its aspirational goals.

Zeki Doruk Erden, Boi Faltings

Artificial intelligence (AI), propelled by advancements in machine learning, has made significant strides in solving complex tasks. However, the current neural network-based paradigm, while effective, is heavily constrained by inherent limitations, primarily a lack of structural organization and a progression of learning that displays undesirable properties. As AI research progresses without a unifying framework, it either tries to patch weaknesses heuristically or draws loosely from biological mechanisms without strong theoretical foundations. Meanwhile, the recent paradigm shift in evolutionary understanding -- driven primarily by evolutionary developmental biology (EDB) -- has been largely overlooked in AI literature, despite a striking analogy between the Modern Synthesis and contemporary machine learning, evident in their shared assumptions, approaches, and limitations upon careful analysis. Consequently, the principles of adaptation from EDB that reshaped our understanding of the evolutionary process can also form the foundation of a unifying conceptual framework for the next design philosophy in AI, going beyond mere inspiration and grounded firmly in biology's first principles. This article provides a detailed overview of the analogy between the Modern Synthesis and modern machine learning, and outlines the core principles of a new AI design paradigm based on insights from EDB. To exemplify our analysis, we also present two learning system designs grounded in specific developmental principles -- regulatory connections, somatic variation and selection, and weak linkage -- that resolve multiple major limitations of contemporary machine learning in an organic manner, while also providing deeper insights into the role of these mechanisms in biological evolution.

Shaobo Cui, Zhijing Jin, Bernhard Schölkopf et al.

Understanding commonsense causality is a unique mark of intelligence for humans. It helps people understand the principles of the real world better and benefits the decision-making process related to causation. For instance, commonsense causality is crucial in judging whether a defendant's action causes the plaintiff's loss in determining legal liability. Despite its significance, a systematic exploration of this topic is notably lacking. Our comprehensive survey bridges this gap by focusing on taxonomies, benchmarks, acquisition methods, qualitative reasoning, and quantitative measurements in commonsense causality, synthesizing insights from over 200 representative articles. Our work aims to provide a systematic overview, update scholars on recent advancements, provide a pragmatic guide for beginners, and highlight promising future research directions in this vital field.

Fengyu Cai, Wanhao Zhou, Fei Mi et al.

Utterance-level intent detection and token-level slot filling are two key tasks for natural language understanding (NLU) in task-oriented systems. Most existing approaches assume that only a single intent exists in an utterance. However, there are often multiple intents within an utterance in real-life scenarios. In this paper, we propose a multi-intent NLU framework, called SLIM, to jointly learn multi-intent detection and slot filling based on BERT. To fully exploit the existing annotation data and capture the interactions between slots and intents, SLIM introduces an explicit slot-intent classifier to learn the many-to-one mapping between slots and intents. Empirical results on three public multi-intent datasets demonstrate (1) the superior performance of SLIM compared to the current state-of-the-art for NLU with multiple intents and (2) the benefits obtained from the slot-intent classifier.

Diana Petrescu, Diego Antognini, Boi Faltings

Recommendations with personalized explanations have been shown to increase user trust and perceived quality and help users make better decisions. Moreover, such explanations allow users to provide feedback by critiquing them. Several algorithms for recommender systems with multi-step critiquing have therefore been developed. However, providing a user-friendly interface based on personalized explanations and critiquing has not been addressed in the last decade. In this paper, we introduce four different web interfaces (available under https://lia.epfl.ch/critiquing/) helping users making decisions and finding their ideal item. We have chosen the hotel recommendation domain as a use case even though our approach is trivially adaptable for other domains. Moreover, our system is model-agnostic (for both recommender systems and critiquing models) allowing a great flexibility and further extensions. Our interfaces are above all a useful tool to help research in recommendation with critiquing. They allow to test such systems on a real use case and also to highlight some limitations of these approaches to find solutions to overcome them.

Diego Antognini, Boi Faltings

Recent studies have shown that providing personalized explanations alongside recommendations increases trust and perceived quality. Furthermore, it gives users an opportunity to refine the recommendations by critiquing parts of the explanations. On one hand, current recommender systems model the recommendation, explanation, and critiquing objectives jointly, but this creates an inherent trade-off between their respective performance. On the other hand, although recent latent linear critiquing approaches are built upon an existing recommender system, they suffer from computational inefficiency at inference due to the objective optimized at each conversation's turn. We address these deficiencies with M&Ms-VAE, a novel variational autoencoder for recommendation and explanation that is based on multimodal modeling assumptions. We train the model under a weak supervision scheme to simulate both fully and partially observed variables. Then, we leverage the generalization ability of a trained M&Ms-VAE model to embed the user preference and the critique separately. Our work's most important innovation is our critiquing module, which is built upon and trained in a self-supervised manner with a simple ranking objective. Experiments on four real-world datasets demonstrate that among state-of-the-art models, our system is the first to dominate or match the performance in terms of recommendation, explanation, and multi-step critiquing. Moreover, M&Ms-VAE processes the critiques up to 25.6x faster than the best baselines. Finally, we show that our model infers coherent joint and cross generation, even under weak supervision, thanks to our multimodal-based modeling and training scheme.

Panayiotis Danassis, Zeki Doruk Erden, Boi Faltings

Can artificial agents benefit from human conventions? Human societies manage to successfully self-organize and resolve the tragedy of the commons in common-pool resources, in spite of the bleak prediction of non-cooperative game theory. On top of that, real-world problems are inherently large-scale and of low observability. One key concept that facilitates human coordination in such settings is the use of conventions. Inspired by human behavior, we investigate the learning dynamics and emergence of temporal conventions, focusing on common-pool resources. Extra emphasis was given in designing a realistic evaluation setting: (a) environment dynamics are modeled on real-world fisheries, (b) we assume decentralized learning, where agents can observe only their own history, and (c) we run large-scale simulations (up to 64 agents). Uncoupled policies and low observability make cooperation hard to achieve; as the number of agents grow, the probability of taking a correct gradient direction decreases exponentially. By introducing an arbitrary common signal (e.g., date, time, or any periodic set of numbers) as a means to couple the learning process, we show that temporal conventions can emerge and agents reach sustainable harvesting strategies. The introduction of the signal consistently improves the social welfare (by 258% on average, up to 3306%), the range of environmental parameters where sustainability can be achieved (by 46% on average, up to 300%), and the convergence speed in low abundance settings (by 13% on average, up to 53%).

Milena Filipovic, Blagoj Mitrevski, Diego Antognini et al.

Recommender systems research tends to evaluate model performance offline and on randomly sampled targets, yet the same systems are later used to predict user behavior sequentially from a fixed point in time. Simulating online recommender system performance is notoriously difficult and the discrepancy between online and offline behaviors is typically not accounted for in offline evaluations. This disparity permits weaknesses to go unnoticed until the model is deployed in a production setting. In this paper, we first demonstrate how omitting temporal context when evaluating recommender system performance leads to false confidence. To overcome this, we postulate that offline evaluation protocols can only model real-life use-cases if they account for temporal context. Next, we propose a training procedure to further embed the temporal context in existing models. We use a multi-objective approach to introduce temporal context into traditionally time-unaware recommender systems and confirm its advantage via the proposed evaluation protocol. Finally, we validate that the Pareto Fronts obtained with the added objective dominate those produced by state-of-the-art models that are only optimized for accuracy on three real-world publicly available datasets. The results show that including our temporal objective can improve recall@20 by up to 20%.

Blagoj Mitrevski, Milena Filipovic, Diego Antognini et al.

Multi-objective gradient methods are becoming the standard for solving multi-objective problems. Among others, they show promising results in developing multi-objective recommender systems with both correlated and conflicting objectives. Classic multi-gradient~descent usually relies on the combination of the gradients, not including the computation of first and second moments of the gradients. This leads to a brittle behavior and misses important areas in the solution space. In this work, we create a multi-objective model-agnostic Adamize method that leverages the benefits of the Adam optimizer in single-objective problems. This corrects and stabilizes~the~gradients of every objective before calculating a common gradient descent vector that optimizes all the objectives simultaneously. We evaluate the benefits of Multi-objective Adamize on two multi-objective recommender systems and for three different objective combinations, both correlated or conflicting. We report significant improvements, measured with three different Pareto front metrics: hypervolume, coverage, and spacing. Finally, we show that the \textit{Adamized} Pareto front strictly dominates the previous one on multiple objective pairs.

Kirtan Padh, Diego Antognini, Emma Lejal Glaude et al.

The goal of fairness in classification is to learn a classifier that does not discriminate against groups of individuals based on sensitive attributes, such as race and gender. One approach to designing fair algorithms is to use relaxations of fairness notions as regularization terms or in a constrained optimization problem. We observe that the hyperbolic tangent function can approximate the indicator function. We leverage this property to define a differentiable relaxation that approximates fairness notions provably better than existing relaxations. In addition, we propose a model-agnostic multi-objective architecture that can simultaneously optimize for multiple fairness notions and multiple sensitive attributes and supports all statistical parity-based notions of fairness. We use our relaxation with the multi-objective architecture to learn fair classifiers. Experiments on public datasets show that our method suffers a significantly lower loss of accuracy than current debiasing algorithms relative to the unconstrained model.

Fei Mi, Boi Faltings

Increasing concerns with privacy have stimulated interests in Session-based Recommendation (SR) using no personal data other than what is observed in the current browser session. Existing methods are evaluated in static settings which rarely occur in real-world applications. To better address the dynamic nature of SR tasks, we study an incremental SR scenario, where new items and preferences appear continuously. We show that existing neural recommenders can be used in incremental SR scenarios with small incremental updates to alleviate computation overhead and catastrophic forgetting. More importantly, we propose a general framework called Memory Augmented Neural model (MAN). MAN augments a base neural recommender with a continuously queried and updated nonparametric memory, and the predictions from the neural and the memory components are combined through another lightweight gating network. We empirically show that MAN is well-suited for the incremental SR task, and it consistently outperforms state-of-the-art neural and nonparametric methods. We analyze the results and demonstrate that it is particularly good at incrementally learning preferences on new and infrequent items.

Nikola Milojkovic, Diego Antognini, Giancarlo Bergamin et al.

Recommender systems need to mirror the complexity of the environment they are applied in. The more we know about what might benefit the user, the more objectives the recommender system has. In addition there may be multiple stakeholders - sellers, buyers, shareholders - in addition to legal and ethical constraints. Simultaneously optimizing for a multitude of objectives, correlated and not correlated, having the same scale or not, has proven difficult so far. We introduce a stochastic multi-gradient descent approach to recommender systems (MGDRec) to solve this problem. We show that this exceeds state-of-the-art methods in traditional objective mixtures, like revenue and recall. Not only that, but through gradient normalization we can combine fundamentally different objectives, having diverse scales, into a single coherent framework. We show that uncorrelated objectives, like the proportion of quality products, can be improved alongside accuracy. Through the use of stochasticity, we avoid the pitfalls of calculating full gradients and provide a clear setting for its applicability.

Boi Faltings, Martin Charles Golumbic

Partitioning a graph using graph separators, and particularly clique separators, are well-known techniques to decompose a graph into smaller units which can be treated independently. It was previously known that the treewidth was bounded above by the sum of the size of the separator plus the treewidth of disjoint components, and this was obtained by the heuristic of filling in all edges of the separator making it into a clique. In this paper, we present a new, tighter upper bound on the treewidth of a graph obtained by only partially filling in the edges of a separator. In particular, the method completes just those pairs of separator vertices that are adjacent to a common component, and indicates a more effective heuristic than filling in the entire separator. We discuss the relevance of this result for combinatorial algorithms and give an example of how the tighter bound can be exploited in the domain of constraint satisfaction problems.

Adam Richardson, Aris Filos-Ratsikas, Boi Faltings

We consider a crowdsourcing data acquisition scenario, such as federated learning, where a Center collects data points from a set of rational Agents, with the aim of training a model. For linear regression models, we show how a payment structure can be designed to incentivize the agents to provide high-quality data as early as possible, based on a characterization of the influence that data points have on the loss function of the model. Our contributions can be summarized as follows: (a) we prove theoretically that this scheme ensures truthful data reporting as a game-theoretic equilibrium and further demonstrate its robustness against mixtures of truthful and heuristic data reports, (b) we design a procedure according to which the influence computation can be efficiently approximated and processed sequentially in batches over time, (c) we develop a theory that allows correcting the difference between the influence and the overall change in loss and (d) we evaluate our approach on real datasets, confirming our theoretical findings.

Naman Goel, Cyril van Schreven, Aris Filos-Ratsikas et al.

Blockchain based systems allow various kinds of financial transactions to be executed in a decentralized manner. However, these systems often rely on a trusted third party (oracle) to get correct information about the real-world events, which trigger the financial transactions. In this paper, we identify two biggest challenges in building decentralized, trustless and transparent oracles. The first challenge is acquiring correct information about the real-world events without relying on a trusted information provider. We show how a peer-consistency incentive mechanism can be used to acquire truthful information from an untrusted and self-interested crowd, even when the crowd has outside incentives to provide wrong information. The second is a system design and implementation challenge. For the first time, we show how to implement a trustless and transparent oracle in Ethereum. We discuss various non-trivial issues that arise in implementing peer-consistency mechanisms in Ethereum, suggest several optimizations to reduce gas cost and provide empirical analysis.

Sankarshan Damle, Boi Faltings, Sujit Gujar

The emergence of e-commerce and e-voting platforms has resulted in the rise in the volume of sensitive information over the Internet. This has resulted in an increased demand for secure and private means of information computation. Towards this, the Yao's Millionaires' problem, i.e., to determine the richer among two millionaires' securely, finds an application. In this work, we present a new solution to the Yao's Millionaires' problem namely, Privacy Preserving Comparison (PPC). We show that PPC achieves this comparison in constant time as well as in one execution. PPC uses semi-honest third parties for the comparison who do not learn any information about the values. Further, we show that PPC is collusion-resistance. To demonstrate the significance of PPC, we present a secure, approximate single-minded combinatorial auction, which we call TPACAS, i.e., Truthful, Privacy-preserving Approximate Combinatorial Auction for Single-minded bidders. We show that TPACAS, unlike previous works, preserves the following privacies relevant to an auction: agent privacy, the identities of the losing bidders must not be revealed to any other agent except the auctioneer (AU), bid privacy, the bid values must be hidden from the other agents as well as the AU and bid-topology privacy, the items for which the agents are bidding must be hidden from the other agents as well as the AU. We demonstrate the practicality of TPACAS through simulations. Lastly, we also look at TPACAS' implementation over a publicly distributed ledger, such as the Ethereum blockchain.

Naman Goel, Boi Faltings

Recent studies have shown that the labels collected from crowdworkers can be discriminatory with respect to sensitive attributes such as gender and race. This raises questions about the suitability of using crowdsourced data for further use, such as for training machine learning algorithms. In this work, we address the problem of fair and diverse data collection from a crowd under budget constraints. We propose a novel algorithm which maximizes the expected accuracy of the collected data, while ensuring that the errors satisfy desired notions of fairness. We provide guarantees on the performance of our algorithm and show that the algorithm performs well in practice through experiments on a real dataset.

Dimitris Chatzopoulos, Sujit Gujar, Boi Faltings et al.

The popularity and applicability of mobile crowdsensing applications are continuously increasing due to the widespread of mobile devices and their sensing and processing capabilities. However, we need to offer appropriate incentives to the mobile users who contribute their resources and preserve their privacy. Blockchain technologies enable semi-anonymous multi-party interactions and can be utilized in crowdsensing applications to maintain the privacy of the mobile users while ensuring first-rate crowdsensed data. In this work, we propose to use blockchain technologies and smart contracts to orchestrate the interactions between mobile crowdsensing providers and mobile users for the case of spatial crowdsensing, where mobile users need to be at specific locations to perform the tasks. Smart contracts, by operating as processes that are executed on the blockchain, are used to preserve users' privacy and make payments. Furthermore, for the assignment of the crowdsensing tasks to the mobile users, we design a truthful, cost-optimal auction that minimizes the payments from the crowdsensing providers to the mobile users. Extensive experimental results show that the proposed privacy preserving auction outperforms state-of-the-art proposals regarding cost by ten times for high numbers of mobile users and tasks.

Fei Mi, Boi Faltings

There has been growing interests in recent years from both practical and research perspectives for session-based recommendation tasks as long-term user profiles do not often exist in many real-life recommendation applications. In this case, recommendations for user's immediate next actions need to be generated based on patterns in anonymous short sessions. An often overlooked aspect is that new items with limited observations arrive continuously in many domains (e.g. news and discussion forums). Therefore, recommendations need to be adaptive to such frequent changes. In this paper, we benchmark a new nonparametric method called context tree (CT) against various state-of-the-art methods on extensive datasets for session-based recommendation task. Apart from the standard static evaluation protocol adopted by previous literatures, we include an adaptive configuration to mimic the situation when new items with limited observations arrives continuously. Our results show that CT outperforms two best-performing approaches (recurrent neural network; heuristic-based nearest neighbor) in majority of the tested configurations and datasets. We analyze reasons for this and demonstrate that it is because of the better adaptation to changes in the domain, as well as the remarkable capability to learn static sequential patterns. Moreover, our running time analysis illustrates the efficiency of using CT as other nonparametric methods.

Naman Goel, Boi Faltings

An important class of game-theoretic incentive mechanisms for eliciting effort from a crowd are the peer based mechanisms, in which workers are paid by matching their answers with one another. The other classic mechanism is to have the workers solve some gold standard tasks and pay them according to their accuracy on gold tasks. This mechanism ensures stronger incentive compatibility than the peer based mechanisms but assigning gold tasks to all workers becomes inefficient at large scale. We propose a novel mechanism that assigns gold tasks to only a few workers and exploits transitivity to derive accuracy of the rest of the workers from their peers' accuracy. We show that the resulting mechanism ensures a dominant notion of incentive compatibility and fairness.

Dimitris Chatzopoulos, Sujit Gujar, Boi Faltings et al.

The popularity of digital currencies, especially cryptocurrencies, has been continuously growing since the appearance of Bitcoin. Bitcoin's security lies in a proof-of-work scheme, which requires high computational resources at the miners. Despite advances in mobile technology, existing cryptocurrencies cannot be maintained by mobile devices due to their low processing capabilities. Mobile devices can only accommodate mobile applications (wallets) that allow users to exchange credits of cryptocurrencies. In this work, we propose LocalCoin, an alternative cryptocurrency that requires minimal computational resources, produces low data traffic and works with off-the-shelf mobile devices. LocalCoin replaces the computational hardness that is at the root of Bitcoin's security with the social hardness of ensuring that all witnesses to a transaction are colluders. Localcoin features (i) a lightweight proof-of-work scheme and (ii) a distributed blockchain. We analyze LocalCoin for double spending for passive and active attacks and prove that under the assumption of sufficient number of users and properly selected tuning parameters the probability of double spending is close to zero. Extensive simulations on real mobility traces, realistic urban settings, and random geometric graphs show that the probability of success of one transaction converges to 1 and the probability of the success of a double spending attempt converges to 0.