Raphaël Koster
YOU?
Author Swipe
View article: Deep mechanism design: Learning social and economic policies for human benefit
Deep mechanism design: Learning social and economic policies for human benefit Open
Human society is coordinated by mechanisms that control how prices are agreed, taxes are set, and electoral votes are tallied. The design of robust and effective mechanisms for human benefit is a core problem in the social, economic, and p…
View article: Tabula rasa agents display emergent in-group behavior
Tabula rasa agents display emergent in-group behavior Open
Theories on group-bias often posit an internal preparedness to bias one’s cognition to favor the in-group (often envisioned as a product of evolution). In contrast, other theories suggest that group-biases can emerge from nonspecialized co…
View article: Modeling human reputation-seeking behavior in a spatio-temporally complex public good provision game
Modeling human reputation-seeking behavior in a spatio-temporally complex public good provision game Open
Multi-agent reinforcement learning algorithms are useful for simulating social behavior in settings that are too complex for other theoretical approaches like game theory. However, they have not yet been empirically supported by laboratory…
View article: Language Agents as Digital Representatives in Collective Decision-Making
Language Agents as Digital Representatives in Collective Decision-Making Open
Consider the process of collective decision-making, in which a group of individuals interactively select a preferred outcome from among a universe of alternatives. In this context, "representation" is the activity of making an individual's…
View article: A theory of appropriateness with applications to generative artificial intelligence
A theory of appropriateness with applications to generative artificial intelligence Open
What is appropriateness? Humans navigate a multi-scale mosaic of interlocking notions of what is appropriate for different situations. We act one way with our friends, another with our family, and yet another in the office. Likewise for AI…
View article: Supported data for manuscript "Can LLM-Augmented autonomous agents cooperate?, An evaluation of their cooperative capabilities through Melting Pot"
Supported data for manuscript "Can LLM-Augmented autonomous agents cooperate?, An evaluation of their cooperative capabilities through Melting Pot" Open
The repository data corresponds partially to the manuscript titled "Can LLM-Augmented Autonomous Agents Cooperate? An Evaluation of Their Cooperative Capabilities through Melting Pot," submitted to IEEE Transactions on Artificial Intellige…
View article: Bridging the digital divide for individuals with intellectual disabilities: Implications for well‐being and inclusion
Bridging the digital divide for individuals with intellectual disabilities: Implications for well‐being and inclusion Open
Background Developments in digital technologies have transformed how people interact with the world, offering employment, education, communication, health benefits and entertainment. Research has shown that not everyone can easily access d…
View article: Using deep reinforcement learning to promote sustainable human behaviour on a common pool resource problem
Using deep reinforcement learning to promote sustainable human behaviour on a common pool resource problem Open
A canonical social dilemma arises when finite resources are allocated to a group of people, who can choose to either reciprocate with interest, or keep the proceeds for themselves. What resource allocation mechanisms will encourage levels …
View article: Beyond the Matrix: Using Multi-Agent-Reinforcement Learning and Behavioral Experiments to Study Social-Ecological Systems
Beyond the Matrix: Using Multi-Agent-Reinforcement Learning and Behavioral Experiments to Study Social-Ecological Systems Open
Social-ecological systems, in which agents interact with each other and their environment are important both for sustainability applications and for understanding how human intelligence functions in context. In such systems, the environmen…
View article: Perception Test: A Diagnostic Benchmark for Multimodal Video Models
Perception Test: A Diagnostic Benchmark for Multimodal Video Models Open
We propose a novel multimodal video benchmark - the Perception Test - to evaluate the perception and reasoning skills of pre-trained multimodal models (e.g. Flamingo, SeViLA, or GPT-4). Compared to existing benchmarks that focus on computa…
View article: A learning agent that acquires social norms from public sanctions in decentralized multi-agent settings
A learning agent that acquires social norms from public sanctions in decentralized multi-agent settings Open
Society is characterized by the presence of a variety of social norms: collective patterns of sanctioning that can prevent miscoordination and free-riding. Inspired by this, we aim to construct learning dynamics where potentially beneficia…
View article: Role of Human-AI Interaction in Selective Prediction
Role of Human-AI Interaction in Selective Prediction Open
Recent work has shown the potential benefit of selective prediction systems that can learn to defer to a human when the predictions of the AI are unreliable, particularly to improve the reliability of AI systems in high-stakes applications…
View article: HCMD-zero: Learning Value Aligned Mechanisms from Data
HCMD-zero: Learning Value Aligned Mechanisms from Data Open
Artificial learning agents are mediating a larger and larger number of interactions among humans, firms, and organizations, and the intersection between mechanism design and machine learning has been heavily investigated in recent years. H…
View article: The Good Shepherd: An Oracle Agent for Mechanism Design
The Good Shepherd: An Oracle Agent for Mechanism Design Open
From social networks to traffic routing, artificial learning agents are playing a central role in modern institutions. We must therefore understand how to leverage these systems to foster outcomes and behaviors that align with our own valu…
View article: Human-centered mechanism design with Democratic AI
Human-centered mechanism design with Democratic AI Open
Building artificial intelligence (AI) that aligns with human values is an unsolved problem. Here, we developed a human-in-the-loop research pipeline called Democratic AI, in which reinforcement learning is used to design a social mechanism…
View article: Scalable Evaluation of Multi-Agent Reinforcement Learning with Melting Pot
Scalable Evaluation of Multi-Agent Reinforcement Learning with Melting Pot Open
Existing evaluation suites for multi-agent reinforcement learning (MARL) do not assess generalization to novel situations as their primary objective (unlike supervised-learning benchmarks). Our contribution, Melting Pot, is a MARL evaluati…
View article: Scalable Evaluation of Multi-Agent Reinforcement Learning with Melting Pot.
Scalable Evaluation of Multi-Agent Reinforcement Learning with Melting Pot. Open
Existing evaluation suites for multi-agent reinforcement learning (MARL) do not assess generalization to novel situations as their primary objective (unlike supervised-learning benchmarks). Our contribution, Melting Pot, is a MARL evaluati…
View article: A learning agent that acquires social norms from public sanctions in decentralized multi-agent settings
A learning agent that acquires social norms from public sanctions in decentralized multi-agent settings Open
Society is characterized by the presence of a variety of social norms: collective patterns of sanctioning that can prevent miscoordination and free-riding. Inspired by this, we aim to construct learning dynamics where potentially beneficia…
View article: A multi-agent reinforcement learning model of reputation and cooperation in human groups
A multi-agent reinforcement learning model of reputation and cooperation in human groups Open
Collective action demands that individuals efficiently coordinate how much, where, and when to cooperate. Laboratory experiments have extensively explored the first part of this process, demonstrating that a variety of social-cognitive mec…
View article: Deep reinforcement learning models the emergent dynamics of human cooperation
Deep reinforcement learning models the emergent dynamics of human cooperation Open
Collective action demands that individuals efficiently coordinate how much, where, and when to cooperate. Laboratory experiments have extensively explored the first part of this process, demonstrating that a variety of social-cognitive mec…
View article: Model-free conventions in multi-agent reinforcement learning with heterogeneous preferences
Model-free conventions in multi-agent reinforcement learning with heterogeneous preferences Open
Game theoretic views of convention generally rest on notions of common knowledge and hyper-rational models of individual behavior. However, decades of work in behavioral economics have questioned the validity of both foundations. Meanwhile…
View article: MEMO: A Deep Network for Flexible Combination of Episodic Memories
MEMO: A Deep Network for Flexible Combination of Episodic Memories Open
Recent research developing neural network architectures with external memory have often used the benchmark bAbI question and answering dataset which provides a challenging number of tasks requiring reasoning. Here we employed a classic ass…
View article: Silly rules improve the capacity of agents to learn stable enforcement and compliance behaviors
Silly rules improve the capacity of agents to learn stable enforcement and compliance behaviors Open
How can societies learn to enforce and comply with social norms? Here we investigate the learning dynamics and emergence of compliance and enforcement of social norms in a foraging game, implemented in a multi-agent reinforcement learning …
View article: Human hippocampal theta oscillations reflect sequential dependencies during spatial planning
Human hippocampal theta oscillations reflect sequential dependencies during spatial planning Open
Movement-related theta oscillations in rodent hippocampus coordinate 'forward sweeps' of location-specific neural activity that could be used to evaluate spatial trajectories online. This raises the possibility that increases in human hipp…
View article: Human hippocampal theta oscillations reflect sequential dependencies during spatial planning
Human hippocampal theta oscillations reflect sequential dependencies during spatial planning Open
Movement-related theta oscillations in rodent hippocampus coordinate ‘forward sweeps’ of location-specific neural activity that could be used to evaluate spatial trajectories online. This raises the possibility that increases in human hipp…