Across disciplines, cooperation is a fundamental research topic. While socially desirable to a population, it often bears a cost to the individual who, in their own self-interest, rationally chooses not to engage in costly cooperation. As such, much work has been done in understanding the biological mechanisms behind cooperation in human and animal populations. In my PhD project, I develop and apply these mechanisms both to artificial multi-agent systems and real social systems. I examine how factors such as agent heterogeneity and different learning algorithms affect not only the level of cooperation within a system, but also the level of fairness in the distribution of payoffs. In previous work, I showed how the effectiveness of the social norm-based mechanism of indirect reciprocity is affected when in-group biased cooperation is present. Beyond my future work on online platforms, I also plan to explore the effects of space, gossip, and partial and subjective observations to widen the potential scope of applications.
论文检索
输入标题、作者或关键词，从 1,048 篇学术成果中精准定位
Due to the complexity of randomness, optimization problems are often modeled to be deterministic to be solvable. Specifically epistemic uncertainty, i.e., uncertainty that is caused due to a lack of knowledge, is not easy to model, let alone easy to subsequently solve. Despite this, taking uncertainty into account is often required for optimization models to produce robust decisions that perform well in practice. We analyze effective existing frameworks, aiming to improve robustness without increasing complexity. Specifically we focus on robustness in decision-focused learning, which is a framework aimed at making context-based predictions for an optimization problem's uncertain parameters that minimize decision error.
In my research, I focus on computationally hard problems in the area of computational social choice. I am interested in the study of input restrictions that guarantee the existence of efficient and scalable algorithms that can be of practical interest.
We investigate the existence and computation of fair and efficient allocations of indivisible chores to agents with additive preferences. We consider the popular envy-based fairness notions of envy-freeness up to one chore (EF1) and the efficiency notion of Pareto-optimality (PO). The existence of an allocation of chores that is simultaneously EF1 and PO is regarded a major open problem in discrete fair division. We show that an EF1 and PO allocation can be computed in polynomial time for certain structured instances. These results comprise the first non-trivial positive results for the problem and reveal insights towards settling the problem in its full generality.
Sequential decision-making under uncertainty in multi-agent environments is a fundamental problem in artificial intelligence. Games serve as a base model for these problems. Finding optimal plans in games that model real-world scenarios necessitates scalable algorithms. In games with perfect information, algorithms that use a combination of search and deep reinforcement learning can scale to arbitrary-sized games and achieve superhuman performance. In games with imperfect information, the situation is more challenging due to the nature of the search. This work aims to develop algorithms that use search but can scale into larger games than currently possible.
Facility location problems have been studied in settings like hospital placement or the competition between stores. In some cases, a central authority coordinates facility placements to optimize metrics like the coverage of an area or emergency response time. In many cases, however, facilities are placed by multiple rational agents to maximize their utility, e.g., the number of clients they attract. In previous research, these games feature simplistic client behavior independent of other clients' strategic choices, e.g., visiting the closest facility. Our goal is to understand what happens if clients also act selfishly, resulting in a two-stage game consisting of strategic facility and client agents. In three recent publications, we investigated such two-stage models for clients that optimize their waiting times. We showed the existence and gave algorithms for (approximate) subgame perfect equilibria, a common extension of Nash equilibria for sequential games. To learn more about this domain, we intend to investigate further natural client behaviors and eventually create a more general model or hierarchy of two-sided facility location games. With this, we aim to make predictions in real-world settings, e.g., the placement of renewable energy infrastructure.
XBRL tagging in financial texts involves categorizing entities into numerous labels, presenting challenges for state-of-the-art models. Financial reports like 10-Q and 10-K, which must be tagged with XBRL according to a taxonomy with thousands of labels. The FNXL dataset exemplifies this with 2,794 labels. Manual tagging is neither scalable nor cost-effective, necessitating automatic annotation methods. Additionally, summarizing long Earnings Call Transcripts (ECTs) is crucial for financial decision-making. The ECTSum dataset highlights challenges in automatic summarization, including a high compression ratio and documents exceeding typical LLM token limits. This study proposes novel methods for both XBRL tagging and ECT summarization.
In recent years, there has been a significant interest in Large Language Models (LLMs) owing to their notable performance in natural language processing (NLP) tasks. However, while their results show promise in mathematical reasoning and software engineering tasks, LLMs have not yet achieved a satisfactory performance level in these domains. In response, current approaches have prioritized scaling up the size of LLMs, necessitating substantial computational resources and data. Our objective, however, is to pursue a different path by developing neurosymbolic language models. We propose to integrate logical and symbolic feedback during the training process, enabling significantly smaller language models to achieve far better reasoning capabilities than the LLMs currently in use.
Causal Graph Modeling with Deep Neural Engines for Strong Abstract Reasoning in Language and Vision
PDF ↗Deep learning (DL) relies on discovering correlation patterns in low-level data and aggregating the information to solve a task. Despite success in a wide variety of applications, ranging from natural language to vision tasks, the learned patterns are often brittle and do not transfer out of the training data distribution (i.e. to different domains). Causality theory proposes methods to discover and estimate cause-effect relationships beyond correlations. Its powerful inference frameworks have been recently highlighted as a potential way to improve the lack of out-of-distribution generalisation in deep neural networks. However, their applications to deep learning problems remain largely under-explored. Our work attempts to bridge this gap and apply causal graphical models to abstract and causal reasoning problems in natural language and vision, requiring strong generalisation abilities beyond correlations. We integrate causal graph modelling methods into deep vision networks and Large Language Models to improve their capacity to perform strong and out-of-distribution reasoning on complex abstract problems.
Smart grid system encompasses large power plants in the wholesale market and retail customers in the tariff market. An electricity broker liaises between the wholesale and tariff markets by procuring electricity from the power plants and selling it to subscribed customers. In our work, we address the prominent challenges in the smart grid system to achieve better efficiency. We discuss the wholesale market, for which we design efficient bidding strategies in periodic double auctions (PDAs), and the tariff market, which includes tariff contract generation strategies and peak demand mitigation strategies. We use the PowerTAC simulator as a test-bed; also utilise these strategies for our autonomous broker, VidyutVanika, which has been proven efficient in the PowerTAC tournaments.
Models for text-to-image synthesis, such as DALL-E 2 and Stable Diffusion, have recently drawn a lot of interest from academia and the general public. These models are capable of producing high-quality images that depict a variety of concepts and styles when conditioned on textual descriptions. However, these models adopt cultural characteristics associated with specific Unicode scripts from their vast amount of training data, which may not be immediately apparent. We show that by simply inserting single non-Latin characters in the textual description, common models reflect cultural biases in their generated images. We analyze this behavior both qualitatively and quantitatively and identify a model’s text encoder as the root cause of the phenomenon. Such behavior can be interpreted as a model feature, offering users a simple way to customize the image generation and reflect their own cultural background. Yet, malicious users or service providers may also try to intentionally bias the image generation. One goal might be to create racist stereotypes by replacing Latin characters with similarly-looking characters from non-Latin scripts, so-called homoglyphs. To mitigate such unnoticed script attacks, we propose a novel homoglyph unlearning method to fine-tune a text encoder, making it robust against homoglyph manipulations.
Hierarchical Decompositions and Termination Analysis for Generalized Planning (Abstract Reprint)
PDF ↗This paper presents new methods for analyzing and evaluating generalized plans that can solve broad classes of related planning problems. Although synthesis and learning of generalized plans has been a longstanding goal in AI, it remains challenging due to fun- damental gaps in methods for analyzing the scope and utility of a given generalized plan. This paper addresses these gaps by developing a new conceptual framework along with proof techniques and algorithmic processes for assessing termination and goal-reachability related properties of generalized plans. We build upon classic results from graph theory to decompose generalized plans into smaller components that are then used to derive hi- erarchical termination arguments. These methods can be used to determine the utility of a given generalized plan, as well as to guide the synthesis and learning processes for generalized plans. We present theoretical as well as empirical results illustrating the scope of this new approach. Our analysis shows that this approach significantly extends the class of generalized plans that can be assessed automatically, thereby reducing barriers in the synthesis and learning of reliable generalized plans.
Conditional Independence (CI) graphs are a type of probabilistic graphical models that are primarily used to gain insights about feature relationships. Each edge represents the partial correlation between the connected features which gives information about their direct dependence. In this survey, we list out different methods and study the advances in techniques developed to recover CI graphs. We cover traditional optimization methods as well as recently developed deep learning architectures along with their recommended implementations . To facilitate wider adoption, we include preliminaries that consolidate associated operations, for example techniques to obtain covariance matrix for mixed datatypes. Keywords: Conditional Independence Graphs, Probabilistic Graphical Models, Graphical Lasso, Deep Learning, Optimization
Performative Ethics From Within the Ivory Tower: How CSPractitioners Uphold Systems of Oppression (Abstract Reprint)
PDF ↗This paper analyzes where Artificial Intelligence (AI) Ethics research fails and breaks down the dangers of well-intentioned, but ultimately performative ethics research. A large majority of AI ethics research is critiqued for lacking a comprehensive analysis of how AI is interconnected with sociological systems of oppression and power. Our work contributes to the handful of research that presents intersectional, Western systems of oppression and power as a framework for examining AI ethics work and the complexities of building less harmful technology; directly connecting technology to named systems such as capitalism and classism, colonialism, racism and white supremacy, patriarchy, and ableism. We then explore current AI ethics rhetoric's effect on the AI ethics domain and AI regulation. In conclusion, we provide an applied example to contextualize intersectional systems of oppression and AI interventions in the U.S. justice system and present actionable steps for AI practitioners to participate in a less performative, critical analysis of AI.
Overfitting in adversarial training has attracted the interest of researchers in the community of artificial intelligence and machine learning in recent years. To address this issue, in this paper we begin by evaluating the defense performances of several calibration methods on various robust models. Our analysis and experiments reveal two intriguing properties: 1) a well-calibrated robust model is decreasing the confidence of robust model; 2) there is a trade-off between the confidences of natural and adversarial images. These new properties offer a straightforward insight into designing a simple but effective regularization, called Self-Residual-Calibration (SRC). The proposed SRC calculates the absolute residual between adversarial and natural logit features corresponding to the ground-truth labels. Furthermore, we utilize the pinball loss to minimize the quantile residual between them, resulting in more robust regularization. Extensive experiments indicate that our SRC can effectively mitigate the overfitting problem while improving the robustness of state-of-the-art models. Importantly, SRC is complementary to various regularization methods. When combined with them, we are capable of achieving the top-rank performance on the AutoAttack benchmark leaderboard.
Partially observable Markov decision processes (POMDPs) provide a flexible representation for real-world decision and control problems. However, POMDPs are notoriously difficult to solve, especially when the state and observation spaces are continuous or hybrid, which is often the case for physical systems. While recent online sampling-based POMDP algorithms that plan with observation likelihood weighting have shown practical effectiveness, a general theory characterizing the approximation error of the particle filtering techniques that these algorithms use has not previously been proposed. Our main contribution is bounding the error between any POMDP and its corresponding finite sample particle belief MDP (PB-MDP) approximation. This fundamental bridge between PB-MDPs and POMDPs allows us to adapt any sampling-based MDP algorithm to a POMDP by solving the corresponding particle belief MDP, thereby extending the convergence guarantees of the MDP algorithm to the POMDP. Practically, this is implemented by using the particle filter belief transition model as the generative model for the MDP solver. While this requires access to the observation density model from the POMDP, it only increases the transition sampling complexity of the MDP solver by a factor of O(C), where C is the number of particles. Thus, when combined with sparse sampling MDP algorithms, this approach can yield algorithms for POMDPs that have no direct theoretical dependence on the size of the state and observation spaces. In addition to our theoretical contribution, we perform five numerical experiments on benchmark POMDPs to demonstrate that a simple MDP algorithm adapted using PB-MDP approximation, Sparse-PFT, achieves performance competitive with other leading continuous observation POMDP solvers.
On Mitigating the Utility-Loss in Differentially Private Learning: A New Perspective by a Geometrically Inspired Kernel Approach (Abstract Reprint)
PDF ↗Privacy-utility tradeoff remains as one of the fundamental issues of differentially private machine learning. This paper introduces a geometrically inspired kernel-based approach to mitigate the accuracy-loss issue in classification. In this approach, a representation of the affine hull of given data points is learned in Reproducing Kernel Hilbert Spaces (RKHS). This leads to a novel distance measure that hides privacy-sensitive information about individual data points and improves the privacy-utility tradeoff via significantly reducing the risk of membership inference attacks. The effectiveness of the approach is demonstrated through experiments on MNIST dataset, Freiburg groceries dataset, and a real biomedical dataset. It is verified that the approach remains computationally practical. The application of the approach to federated learning is considered and it is observed that the accuracy-loss due to data being distributed is either marginal or not significantly high.
SensorSCAN: Self-supervised learning and deep clustering for fault diagnosis in chemical processes (Abstract Reprint)
PDF ↗Modern industrial facilities generate large volumes of raw sensor data during the production process. This data is used to monitor and control the processes and can be analyzed to detect and predict process abnormalities. Typically, the data has to be annotated by experts in order to be used in predictive modeling. However, manual annotation of large amounts of data can be difficult in industrial settings. In this paper, we propose SensorSCAN, a novel method for unsupervised fault detection and diagnosis, designed for industrial chemical process monitoring. We demonstrate our model's performance on two publicly available datasets of the Tennessee Eastman Process with various faults. The results show that our method significantly outperforms existing approaches (+0.2-0.3 TPR for a fixed FPR) and effectively detects most of the process faults without expert annotation. Moreover, we show that the model fine-tuned on a small fraction of labeled data nearly reaches the performance of a SOTA model trained on the full dataset. We also demonstrate that our method is suitable for real-world applications where the number of faults is not known in advance. The code is available at https://github.com/AIRI-Institute/sensorscan
Learning first-order logic programs from relational facts yields intuitive insights into the data. Inductive logic programming (ILP) models are effective in learning first-order logic programs from observed relational data. Symbolic ILP models support rule learning in a data-ecient manner. However, symbolic ILP models are not robust to learn from noisy data. Neuro-symbolic ILP models utilize neural networks to learn logic programs in a differentiable manner which improves the robustness of ILP models. However, most neuro-symbolic methods need a strong language bias to learn logic programs, which reduces the usability and flexibility of ILP models and limits the logic program formats. In addition, most neuro-symbolic ILP methods cannot learn logic programs effectively from both small-size datasets and large-size datasets such as knowledge graphs. In the paper, we introduce a novel differentiable ILP model called differentiable first-order rule learner (DFORL), which is scalable to learn rules from both smaller and larger datasets. Besides, DFORL only needs the number of variables in the learned logic programs as input. Hence, DFORL is easy to use and does not need a strong language bias. We demonstrate that DFORL can perform well on several standard ILP datasets, knowledge graphs, and probabilistic relation facts and outperform several well-known differentiable ILP models. Experimental results indicate that DFORL is a precise, robust, scalable, and computationally cheap differentiable ILP model.
Social dilemmasare situations of inter-dependent decision making in which individualrationality can lead to outcomes with poor social qualities. The ubiquity of social dilem-mas in social, biological, and computational systems has generated substantial researchacross these diverse disciplines into the study of mechanisms for avoiding deficient outcomes by promoting and maintaining mutual cooperation. Much of this research is focused on studying how individuals faced with a dilemma can learn to cooperate by adapting their behaviours according to their past experience. In particular, three types of learning approaches have been studied: evolutionary game-theoretic learning, reinforcement learning, and best-response learning. This article is a comprehensive integrated survey of these learning approaches in the context of dilemma games. We formally introduce dilemma games and their inherent challenges. We then outline the three learning approaches and, for eachapproach, provide a survey of the solutions proposed for dilemma resolution. Finally, we provide a comparative summary and discuss directions in which further research is needed.