Kate Larson
YOU?
Author Swipe
View article: What Voting Rules Actually Do: A Data-Driven Analysis of Multi-Winner Voting
What Voting Rules Actually Do: A Data-Driven Analysis of Multi-Winner Voting Open
Committee-selection problems arise in many contexts and applications, and there has been increasing interest within the social choice research community on identifying which properties are satisfied by different multi-winner voting rules. …
View article: Multi-Agent Risks from Advanced AI
Multi-Agent Risks from Advanced AI Open
The rapid development of advanced AI agents and the imminent deployment of many instances of these agents will give rise to multi-agent systems of unprecedented complexity. These systems pose novel and under-explored risks. In this report,…
View article: Jackpot! Alignment as a Maximal Lottery
Jackpot! Alignment as a Maximal Lottery Open
Reinforcement Learning from Human Feedback (RLHF), the standard for aligning Large Language Models (LLMs) with human values, is known to fail to satisfy properties that are intuitively desirable, such as respecting the preferences of the m…
View article: Imagining and building wise machines: The centrality of AI metacognition
Imagining and building wise machines: The centrality of AI metacognition Open
Although AI has become increasingly smart, its wisdom has not kept pace. In this article, we examine what is known about human wisdom and sketch a vision of its AI counterpart. We analyze human wisdom as a set of strategies for solving int…
View article: Soft Condorcet Optimization for Ranking of General Agents
Soft Condorcet Optimization for Ranking of General Agents Open
Driving progress of AI models and agents requires comparing their performance on standardized benchmarks; for general agents, individual performances must be aggregated across a potentially wide variety of different tasks. In this paper, w…
View article: Democratizing Reward Design for Personal and Representative Value-Alignment
Democratizing Reward Design for Personal and Representative Value-Alignment Open
Aligning AI agents with human values is challenging due to diverse and subjective notions of values. Standard alignment methods often aggregate crowd feedback, which can result in the suppression of unique or minority preferences. We intro…
View article: Liquid Ensemble Selection for Continual Learning
Liquid Ensemble Selection for Continual Learning Open
Continual learning aims to enable machine learning models to continually learn from a shifting data distribution without forgetting what has already been learned. Such shifting distributions can be broken into disjoint subsets of related e…
View article: Unraveling the Dilemma of AI Errors: Exploring the Effectiveness of Human and Machine Explanations for Large Language Models
Unraveling the Dilemma of AI Errors: Exploring the Effectiveness of Human and Machine Explanations for Large Language Models Open
The field of eXplainable artificial intelligence (XAI) has produced a plethora of methods (e.g., saliency-maps) to gain insight into artificial intelligence (AI) models, and has exploded with the rise of deep learning (DL). However, human-…
View article: Approximating the Core via Iterative Coalition Sampling
Approximating the Core via Iterative Coalition Sampling Open
The core is a central solution concept in cooperative game theory, defined as the set of feasible allocations or payments such that no subset of agents has incentive to break away and form their own subgroup or coalition. However, it has l…
View article: Liquid Democracy for Low-Cost Ensemble Pruning
Liquid Democracy for Low-Cost Ensemble Pruning Open
We argue that there is a strong connection between ensemble learning and a delegative voting paradigm -- liquid democracy -- that can be leveraged to reduce ensemble training costs. We present an incremental training procedure that identif…
View article: Evaluating Agents using Social Choice Theory
Evaluating Agents using Social Choice Theory Open
We argue that many general evaluation problems can be viewed through the lens of voting theory. Each task is interpreted as a separate voter, which requires only ordinal rankings or pairwise comparisons of agents to produce an overall eval…
View article: Deliberation and Voting in Approval-Based Multi-Winner Elections
Deliberation and Voting in Approval-Based Multi-Winner Elections Open
Citizen-focused democratic processes where participants deliberate on alternatives and then vote to make the final decision are increasingly popular today. While the computational social choice literature has extensively investigated votin…
View article: Multi-Agent Advisor Q-Learning (Extended Abstract)
Multi-Agent Advisor Q-Learning (Extended Abstract) Open
In the last decade, there have been significant advances in multi-agent reinforcement learning (MARL) but there are still numerous challenges, such as high sample complexity and slow convergence to stable policies, that need to be overcome…
View article: Towards a Better Understanding of Learning with Multiagent Teams
Towards a Better Understanding of Learning with Multiagent Teams Open
While it has long been recognized that a team of individual learning agents can be greater than the sum of its parts, recent work has shown that larger teams are not necessarily more effective than smaller ones. In this paper, we study why…
View article: Towards a Better Understanding of Learning with Multiagent Teams
Towards a Better Understanding of Learning with Multiagent Teams Open
While it has long been recognized that a team of individual learning agents can be greater than the sum of its parts, recent work has shown that larger teams are not necessarily more effective than smaller ones. In this paper, we study why…
View article: Deliberation and Voting in Approval-Based Multi-Winner Elections
Deliberation and Voting in Approval-Based Multi-Winner Elections Open
Citizen-focused democratic processes where participants deliberate on alternatives and then vote to make the final decision are increasingly popular today. While the computational social choice literature has extensively investigated votin…
View article: Revealed Multi-Objective Utility Aggregation in Human Driving
Revealed Multi-Objective Utility Aggregation in Human Driving Open
A central design problem in game theoretic analysis is the estimation of the players' utilities. In many real-world interactive situations of human decision making, including human driving, the utilities are multi-objective in nature; ther…
View article: Combining Deep Reinforcement Learning and Search with Generative Models for Game-Theoretic Opponent Modeling
Combining Deep Reinforcement Learning and Search with Generative Models for Game-Theoretic Opponent Modeling Open
Opponent modeling methods typically involve two crucial steps: building a belief distribution over opponents' strategies, and exploiting this opponent model by playing a best response. However, existing approaches typically require domain-…
View article: Learning from Multiple Independent Advisors in Multi-agent Reinforcement Learning
Learning from Multiple Independent Advisors in Multi-agent Reinforcement Learning Open
Multi-agent reinforcement learning typically suffers from the problem of sample inefficiency, where learning suitable policies involves the use of many data samples. Learning from external demonstrators is a possible solution that mitigate…
View article: Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments
Developing, Evaluating and Scaling Learning Agents in Multi-Agent Environments Open
The Game Theory & Multi-Agent team at DeepMind studies several aspects of multi-agent learning ranging from computing approximations to fundamental concepts in game theory to simulating social dilemmas in rich spatial environments and trai…
View article: Exploring the Benefits of Teams in Multiagent Learning
Exploring the Benefits of Teams in Multiagent Learning Open
For problems requiring cooperation, many multiagent systems implement solutions among either individual agents or across an entire population towards a common goal. Multiagent teams are primarily studied when in conflict; however, organiza…
View article: How Should We Vote? A Comparison of Voting Systems within Social Networks
How Should We Vote? A Comparison of Voting Systems within Social Networks Open
Voting is a crucial methodology for eliciting and combining agents' preferences and information across many applications. Just as there are numerous voting rules exhibiting different properties, we also see many different voting systems. I…
View article: Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior
Generalized Dynamic Cognitive Hierarchy Models for Strategic Driving Behavior Open
While there has been an increasing focus on the use of game theoretic models for autonomous driving, empirical evidence shows that there are still open questions around dealing with the challenges of common knowledge assumptions as well as…
View article: Multi-Agent Advisor Q-Learning
Multi-Agent Advisor Q-Learning Open
In the last decade, there have been significant advances in multi-agent reinforcement learning (MARL) but there are still numerous challenges, such as high sample complexity and slow convergence to stable policies, that need to be overcome…
View article: Exploring the Benefits of Teams in Multiagent Learning
Exploring the Benefits of Teams in Multiagent Learning Open
For problems requiring cooperation, many multiagent systems implement solutions among either individual agents or across an entire population towards a common goal. Multiagent teams are primarily studied when in conflict; however, organiza…
View article: The Importance of Credo in Multiagent Learning
The Importance of Credo in Multiagent Learning Open
We propose a model for multi-objective optimization, a credo, for agents in a system that are configured into multiple groups (i.e., teams). Our model of credo regulates how agents optimize their behavior for the groups they belong to. We …
View article: Multi-Agent Advisor Q-Learning
Multi-Agent Advisor Q-Learning Open
In the last decade, there have been significant advances in multi-agent reinforcement learning (MARL) but there are still numerous challenges, such as high sample complexity and slow convergence to stable policies, that need to be overcome…
View article: A taxonomy of strategic human interactions in traffic conflicts
A taxonomy of strategic human interactions in traffic conflicts Open
In order to enable autonomous vehicles (AV) to navigate busy traffic situations, in recent years there has been a focus on game-theoretic models for strategic behavior planning in AVs. However, a lack of common taxonomy impedes a broader u…