DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
hub
Multi-Goal Reinforcement Learning: Challenging Robotics Environments and Request for Research
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
abstract
The purpose of this technical report is two-fold. First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware. The tasks include pushing, sliding and pick & place with a Fetch robotic arm as well as in-hand object manipulation with a Shadow Dexterous Hand. All tasks have sparse binary rewards and follow a Multi-Goal Reinforcement Learning (RL) framework in which an agent is told what to do using an additional input. The second part of the paper presents a set of concrete research ideas for improving RL algorithms, most of which are related to Multi-Goal RL and Hindsight Experience Replay.
hub tools
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
Reinforcement learning models trained only in simulation using automatic domain randomization solve Rubik's cube with a real robot hand.
A new marginalized reparameterization estimator allows low-variance training of mixture policies in entropy-regularized actor-critic algorithms, matching or exceeding Gaussian policy performance in several continuous control benchmarks.
Ada-Diffuser is a causal diffusion model that jointly learns observed interaction structure and underlying latent dynamics from minimal observations for adaptive planning and policy learning.
Introduces RAPCs and a contraction Bellman operator for cost-optimal policies that satisfy probabilistic reach-avoid specifications in stochastic MDPs, with almost-sure convergence to local optima.
Model-free RL learns optimal strategies in stochastic games for LTL specs by constructing a product with DPA and assigning rewards/discounts from acceptance conditions.
Disentangled Skill Embeddings (DSE) is a variational inference framework for multi-task RL using shared parameters and task-specific latent embeddings for generalization to unseen conditions and as skills in hierarchical RL.
QHyer replaces return-to-go with a state-conditioned Q-estimator and adds a gated hybrid attention-mamba backbone to achieve state-of-the-art performance in offline goal-conditioned RL on both Markovian and non-Markovian datasets.
Trajectory-based data augmentation exploits geometric relationships between rewards, values, and logging policies to enable effective offline RL from few suboptimal trajectories.
Robots outperform constrained human demonstrations by inferring state-only rewards from demos and using temporal interpolation to label and explore better trajectories, achieving 10x faster task completion on a real robotic arm than behavioral cloning.
D2AC combines a diffusion actor with a distributional critic via fused distributional RL and clipped double Q-learning to reach state-of-the-art results on 18 hard control benchmarks including Humanoid, Dog, and Shadow Hand.
SSE improves long-horizon goal-conditioned RL by using failure and partial-success transitions to identify unreliable subgoals, streamline high-level planning, and outperform prior hierarchical methods on benchmarks.
A Goal-Conditioned Decision Transformer is adapted for offline multi-goal RL and shown to outperform online baselines on a new Franka Emika Panda dataset.
Middle-mile logistics is cast as a multi-object goal-conditioned MDP and solved by combining graph neural networks with model-free RL via extraction of small feature graphs.
citing papers explorer
-
Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling
DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
-
Solving Rubik's Cube with a Robot Hand
Reinforcement learning models trained only in simulation using automatic domain randomization solve Rubik's cube with a real robot hand.
-
Revisiting Mixture Policies in Entropy-Regularized Actor-Critic
A new marginalized reparameterization estimator allows low-variance training of mixture policies in entropy-regularized actor-critic algorithms, matching or exceeding Gaussian policy performance in several continuous control benchmarks.
-
Ada-Diffuser: Latent-Aware Adaptive Diffusion for Decision-Making
Ada-Diffuser is a causal diffusion model that jointly learns observed interaction structure and underlying latent dynamics from minimal observations for adaptive planning and policy learning.
-
Stochastic Minimum-Cost Reach-Avoid Reinforcement Learning
Introduces RAPCs and a contraction Bellman operator for cost-optimal policies that satisfy probabilistic reach-avoid specifications in stochastic MDPs, with almost-sure convergence to local optima.
-
Learning Optimal Strategies for Temporal Tasks in Stochastic Games
Model-free RL learns optimal strategies in stochastic games for LTL specs by constructing a product with DPA and assigning rewards/discounts from acceptance conditions.
-
Disentangled Skill Embeddings for Reinforcement Learning
Disentangled Skill Embeddings (DSE) is a variational inference framework for multi-task RL using shared parameters and task-specific latent embeddings for generalization to unseen conditions and as skills in hierarchical RL.
-
QHyer: Q-conditioned Hybrid Attention-mamba Transformer for Offline Goal-conditioned RL
QHyer replaces return-to-go with a state-conditioned Q-estimator and adds a gated hybrid attention-mamba backbone to achieve state-of-the-art performance in offline goal-conditioned RL on both Markovian and non-Markovian datasets.
-
Trajectory-Level Data Augmentation for Offline Reinforcement Learning
Trajectory-based data augmentation exploits geometric relationships between rewards, values, and logging policies to enable effective offline RL from few suboptimal trajectories.
-
When a Robot is More Capable than a Human: Learning from Constrained Demonstrators
Robots outperform constrained human demonstrations by inferring state-only rewards from demos and using temporal interpolation to label and explore better trajectories, achieving 10x faster task completion on a real robotic arm than behavioral cloning.
-
D2 Actor Critic: Diffusion Actor Meets Distributional Critic
D2AC combines a diffusion actor with a distributional critic via fused distributional RL and clipped double Q-learning to reach state-of-the-art results on 18 hard control benchmarks including Humanoid, Dog, and Shadow Hand.
-
Strict Subgoal Execution: Reliable Long-Horizon Planning in Hierarchical Reinforcement Learning
SSE improves long-horizon goal-conditioned RL by using failure and partial-success transitions to identify unreliable subgoals, streamline high-level planning, and outperform prior hierarchical methods on benchmarks.
-
Goal-Conditioned Decision Transformer for Multi-Goal Offline Reinforcement Learning
A Goal-Conditioned Decision Transformer is adapted for offline multi-goal RL and shown to outperform online baselines on a new Franka Emika Panda dataset.
-
Middle-mile logistics through the lens of goal-conditioned reinforcement learning
Middle-mile logistics is cast as a multi-object goal-conditioned MDP and solved by combining graph neural networks with model-free RL via extraction of small feature graphs.