65 citations · 364 across the 38 of their papers we have counts for
27 papers · 1 filter
Comparing the Efficacy of Fine-Tuning and Meta-Learning for Few-Shot Policy Imitation
Massimiliano Patacchiola, Mingfei Sun, Katja Hofmann +1
In this paper we explore few-shot imitation learning for control problems, which involves learning to imitate a target policy by accessing a limited set of offline rollouts. This s…
Learning Safety Constraints from Demonstrations with Unknown Rewards
David Lindner, Xin Chen, Sebastian Tschiatschek +2
We propose Convex Constraint Learning for Reinforcement Learning (CoCoRL), a novel approach for inferring shared constraints in a Constrained Markov Decision Process (CMDP) from a…
Trust-Region-Free Policy Optimization for Stochastic Policies
Mingfei Sun, Benjamin Ellis, Anuj Mahajan +3
Trust Region Policy Optimization (TRPO) is an iterative method that simultaneously maximizes a surrogate objective and enforces a trust region constraint over consecutive policies…
UniMASK: Unified Inference in Sequential Decision Problems
Micah Carroll, Orr Paradise, Jessy Lin +8
Randomly masking and predicting word tokens has been a successful approach in pre-training language models for a variety of downstream tasks. In this work, we observe that the same…
Interactively Learning Preference Constraints in Linear Bandits
David Lindner, Sebastian Tschiatschek, Katja Hofmann +1
We study sequential decision-making with known rewards and unknown constraints, motivated by situations where the constraints represent expensive-to-evaluate human preferences, suc…
Towards Flexible Inference in Sequential Decision Problems via Bidirectional Transformers
Micah Carroll, Jessy Lin, Orr Paradise +8
Randomly masking and predicting word tokens has been a successful approach in pre-training language models for a variety of downstream tasks. In this work, we observe that the same…
Trust Region Bounds for Decentralized PPO Under Non-stationarity
Mingfei Sun, Sam Devlin, Jacob Beck +2
We present trust region bounds for optimizing decentralized policies in cooperative Multi-Agent Reinforcement Learning (MARL), which holds even when the transition dynamics are non…
You May Not Need Ratio Clipping in PPO
Mingfei Sun, Vitaly Kurin, Guoqing Liu +4
Proximal Policy Optimization (PPO) methods learn a policy by iteratively performing multiple mini-batch optimization epochs of a surrogate objective with one set of sampled data. R…
Deterministic and Discriminative Imitation (D2-Imitation): Revisiting Adversarial Imitation for Sample Efficiency
Mingfei Sun, Sam Devlin, Katja Hofmann +1
Sample efficiency is crucial for imitation learning methods to be applicable in real-world applications. Many studies improve sample efficiency by extending adversarial imitation t…
SocialAI: Benchmarking Socio-Cognitive Abilities in Deep Reinforcement Learning Agents
Grgur Kovač, Rémy Portelas, Katja Hofmann +1
Building embodied autonomous agents capable of participating in social interactions with humans is one of the main challenges in AI. Within the Deep Reinforcement Learning (DRL) fi…
Strategically Efficient Exploration in Competitive Multi-agent Reinforcement Learning
Robert Loftin, Aadirupa Saha, Sam Devlin +1
High sample complexity remains a barrier to the application of reinforcement learning (RL), particularly in multi-agent systems. A large body of work has demonstrated that explorat…
SoftDICE for Imitation Learning: Rethinking Off-policy Distribution Matching
Mingfei Sun, Anuj Mahajan, Katja Hofmann +1
We present SoftDICE, which achieves state-of-the-art performance for imitation learning. SoftDICE fixes several key problems in ValueDICE, an off-policy distribution matching appro…
SocialAI 0.1: Towards a Benchmark to Stimulate Research on Socio-Cognitive Abilities in Deep Reinforcement Learning Agents
Grgur Kovač, Rémy Portelas, Katja Hofmann +1
Building embodied autonomous agents capable of participating in social interactions with humans is one of the main challenges in AI. This problem motivated many research directions…
TeachMyAgent: a Benchmark for Automatic Curriculum Learning in Deep RL
Clément Romac, Rémy Portelas, Katja Hofmann +1
Training autonomous agents able to generalize to multiple tasks is a key target of Deep Reinforcement Learning (DRL) research. In parallel to improving DRL algorithms themselves, A…
Evaluating the Robustness of Collaborative Agents
Paul Knott, Micah Carroll, Sam Devlin +4
In order for agents trained by deep reinforcement learning to work alongside humans in realistic settings, we will need to ensure that the agents are \emph{robust}. Since the real…
Deep Interactive Bayesian Reinforcement Learning via Meta-Learning
Luisa Zintgraf, Sam Devlin, Kamil Ciosek +2
Agents that interact with other agents often do not know a priori what the other agents' strategies are, but have to maximise their own online return while interacting with and lea…
Meta Automatic Curriculum Learning
Rémy Portelas, Clément Romac, Katja Hofmann +1
A major challenge in the Deep RL (DRL) community is to train agents able to generalize their control policy over situations never seen in training. Training on diverse tasks has be…
Exploration in Approximate Hyper-State Space for Meta Reinforcement Learning
Luisa Zintgraf, Leo Feng, Cong Lu +4
To rapidly learn a new task, it is often essential for agents to explore efficiently -- especially when performance matters from the first timestep. One way to learn such behaviour…
Analytic Manifold Learning: Unifying and Evaluating Representations for Continuous Control
Rika Antonova, Maksim Maydanskiy, Danica Kragic +2
We address the problem of learning reusable state representations from streaming high-dimensional observations. This is important for areas like Reinforcement Learning (RL), which…
Guaranteeing Reproducibility in Deep Learning Competitions
Brandon Houghton, Stephanie Milani, Nicholay Topin +5
To encourage the development of methods with reproducible and robust training behavior, we propose a challenge paradigm where competitors are evaluated directly on the performance…
Trying AGAIN instead of Trying Longer: Prior Learning for Automatic Curriculum Learning
Rémy Portelas, Katja Hofmann, Pierre-Yves Oudeyer
A major challenge in the Deep RL (DRL) community is to train agents able to generalize over unseen situations, which is often approached by training them on a diversity of tasks (o…
SpatialSim: Recognizing Spatial Configurations of Objects with Graph Neural Networks
Laetitia Teodorescu, Katja Hofmann, Pierre-Yves Oudeyer
Recognizing precise geometrical configurations of groups of objects is a key capability of human spatial cognition, yet little studied in the deep learning literature so far. In pa…
Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck
Maximilian Igl, Kamil Ciosek, Yingzhen Li +4
The ability for policies to generalize to new environments is key to the broad application of RL agents. A promising approach to prevent an agent's policy from overfitting to a lim…
Teacher algorithms for curriculum learning of Deep RL in continuously parameterized environments
Rémy Portelas, Cédric Colas, Katja Hofmann +1
We consider the problem of how a teacher algorithm can enable an unknown Deep Reinforcement Learning (DRL) student to become good at a skill over a wide range of diverse environmen…