Search Results for author:
Found 19 papers, 11 papers with code
Multi-Agent Risks from Advanced AI
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.
BAMDP Shaping: a Unified Theoretical Framework for Intrinsic Motivation and Reward Shaping
Intrinsic motivation (IM) and reward shaping are common methods for guiding the exploration of reinforcement learning (RL) agents by adding pseudo-rewards.
The Benefits of Power Regularization in Cooperative Reinforcement Learning
Cooperative Multi-Agent Reinforcement Learning (MARL) algorithms, trained only to optimize task reward, can lead to a concentration of power where the failure or adversarial intent of a single agent could decimate the reward of every agent in the system.
Multi-agent Reinforcement Learning
reinforcement-learning
+1
Open-Endedness is Essential for Artificial Superhuman Intelligence
In this position paper, we argue that the ingredients are now in place to achieve openendedness in AI systems with respect to a human observer.
Genie: Generative Interactive Environments
We introduce Genie, the first generative interactive environment trained in an unsupervised manner from unlabelled Internet videos.
Refining Minimax Regret for Unsupervised Environment Design
In this work, we introduce Bayesian level-perfect MMR (BLP), a refinement of the minimax regret objective that overcomes this limitation.
minimax: Efficient Baselines for Autocurricula in JAX
This compute requirement is a major obstacle to rapid innovation for the field.
Stabilizing Unsupervised Environment Design with a Learned Adversary
As a result, we make it possible for PAIRED to match or exceed state-of-the-art methods, producing robust agents in several established challenging procedurally-generated environments, including a partially-observed maze navigation task and a continuous-control car racing environment.
Who Needs to Know? Minimal Knowledge for Optimal Coordination
We apply this algorithm to analyze the strategically relevant information for tasks in both a standard and a partially observable version of the Overcooked environment.
MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning
Open-ended learning methods that automatically generate a curriculum of increasingly challenging tasks serve as a promising avenue toward generally capable reinforcement learning agents.
Grounding Aleatoric Uncertainty for Unsupervised Environment Design
Problematically, in partially-observable or stochastic settings, optimal policies may depend on the ground-truth distribution over aleatoric parameters of the environment in the intended deployment setting, while curriculum learning necessarily shifts the training distribution.
Evolving Curricula with Regret-Based Environment Design
Our approach, which we call Adversarially Compounding Complexity by Editing Levels (ACCEL), seeks to constantly produce levels at the frontier of an agent's capabilities, resulting in curricula that start simple but become increasingly complex.
Replay-Guided Adversarial Environment Design
Furthermore, our theory suggests a highly counterintuitive improvement to PLR: by stopping the agent from updating its policy on uncurated levels (training on less data), we can improve the convergence to Nash equilibria.
A New Formalism, Method and Open Issues for Zero-Shot Coordination
We introduce an extension of the algorithm, other-play with tie-breaking, and prove that it is optimal in the LFC problem and an equilibrium in the LFC game.
Improving Social Welfare While Preserving Autonomy via a Pareto Mediator
Machine learning algorithms often make decisions on behalf of agents with varied and sometimes conflicting interests.
Accumulating Risk Capital Through Investing in Cooperation
Recent work on promoting cooperation in multi-agent learning has resulted in many methods which successfully promote cooperation at the cost of becoming more vulnerable to exploitation by malicious actors.
Emergent Complexity and Zero-shot Transfer via Unsupervised Environment Design
We call our technique Protagonist Antagonist Induced Regret Environment Design (PAIRED).
Quantifying Differences in Reward Functions
However, this method cannot distinguish between the learned reward function failing to reflect user preferences and the policy optimization process failing to optimize the learned reward.
Adversarial Policies: Attacking Deep Reinforcement Learning
Deep reinforcement learning (RL) policies are known to be vulnerable to adversarial perturbations to their observations, similar to adversarial examples for classifiers.
