REVIEW 25 cited by
Mastering Visual Continuous Control: Improved Data-Augmented Reinforcement Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We present DrQ-v2, a model-free reinforcement learning (RL) algorithm for visual continuous control. DrQ-v2 builds on DrQ, an off-policy actor-critic approach that uses data augmentation to learn directly from pixels. We introduce several improvements that yield state-of-the-art results on the DeepMind Control Suite. Notably, DrQ-v2 is able to solve complex humanoid locomotion tasks directly from pixel observations, previously unattained by model-free RL. DrQ-v2 is conceptually simple, easy to implement, and provides significantly better computational footprint compared to prior work, with the majority of tasks taking just 8 hours to train on a single GPU. Finally, we publicly release DrQ-v2's implementation to provide RL practitioners with a strong and computationally efficient baseline.
Forward citations
Cited by 25 Pith papers
-
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.
-
Operator-Guided Invariance Learning for Continuous Reinforcement Learning
VPSD-RL discovers exact and approximate value-preserving Lie-group operators in continuous RL to stabilize learning via transition augmentation and consistency regularization.
-
Agent-Centric Observation Adaptation for Robust Visual Control under Dynamic Perturbations
ACO-MoE employs agent-centric mixture-of-experts to decouple task-relevant features from dynamic visual perturbations in RL, recovering 95.3% of clean performance on the new VDCS benchmark.
-
Agent-Centric Observation Adaptation for Robust Visual Control under Dynamic Perturbations
ACO-MoE recovers 95.3% of clean-input performance in visual control tasks under Markov-switching corruptions by routing restoration experts and anchoring representations to clean foreground masks.
-
Mastering Diverse Domains through World Models
DreamerV3 uses world models and robustness techniques to solve over 150 tasks across domains with a single configuration, including Minecraft diamond collection from scratch.
-
OmniTacTune: Policy-Agnostic Real-World RL for Tactile Residual Adaptation of Visual Policies
A policy-agnostic two-stage real-world RL method learns tactile residual corrections on frozen visual policies, lifting contact-rich task success from 5–40% to 85–100% in under 80 minutes.
-
Rank-Then-Act: Reward-Free Control from Frame-Order Progress
RTA trains a VLM as a progress ordinal scorer via GRPO on shuffled expert frames and uses Spearman rank correlation with temporal indices as a bounded RL reward, matching or exceeding prior video reward methods on dis...
-
RARM: Confidence-Gated Progress Reward Modeling for RL in Manipulation
RARM is a lightweight visual comparator trained once on general videos that supplies dense progress rewards to RL by matching rollout clips to a reference demonstration and gating rewards on match confidence.
-
FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control
FlashSAC scales up Soft Actor-Critic with fewer updates, larger models, higher data throughput, and norm bounds to deliver faster, more stable training than PPO on high-dimensional robot control tasks across dozens of...
-
FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control
FlashSAC improves training speed and final performance of off-policy RL on high-dimensional robot tasks by reducing update frequency, increasing model scale, and bounding norms to limit critic error accumulation.
-
Error Amplification Limits ANN-to-SNN Conversion in Continuous Control
Temporally correlated action errors, amplified by closed-loop dynamics, explain ANN-to-SNN conversion failures in continuous control, and cross-step residual potential initialization mitigates them.
-
Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach
A sample-efficient SAC-based method with replay-ratio resets, offline data bootstrapping, and expert-driven fine-tuning produces a goalkeeper that outperforms the built-in AI in EA SPORTS FC 25.
-
TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
TimeRewarder derives step-wise progress rewards from frame-wise temporal distances in passive videos and uses them to guide RL, achieving high success rates on Meta-World tasks with fewer interactions than prior metho...
-
DeGuV: Depth-Guided Visual Reinforcement Learning for Generalization and Interpretability in Manipulation
Depth-guided masking improves visual RL generalization, sample efficiency, and interpretability on manipulation tasks.
-
HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model
HybridVLA unifies diffusion and autoregression in a single VLA model via collaborative training and ensemble to raise robot manipulation success rates by 14% in simulation and 19% in real-world tasks.
-
Differential Amplifier-Inspired AmpAttention for Multi-View Robotic Manipulation
AmpAttention and RVAF raise multi-view robotic manipulation success and cut training time by suppressing attention noise with a differential-amplifier-style mechanism plus a CMRR loss.
-
What Matters for Simulation to Online Reinforcement Learning on Real Robots
Sim-to-online RL on three real robots is stabilized by retaining data, warm-starting the replay buffer, and using asymmetric actor-critic updates with a low actor learning rate.
-
ReinforceGen: Hybrid Skill Policies with Automated Data Generation and Reinforcement Learning
ReinforceGen uses imitation learning, RL fine-tuning of skill policies, and real-time pose replanning to reach over 80% success on five long-horizon Robosuite tasks from only 10 human demonstrations.
-
RoDyn: Taming Interactive Robot-Dynamic 2.5D World Model for Robotic Manipulation
Abstract describes RoDyn but full text describes iMoWM; the record is internally inconsistent and the headline claims are absent from the body.
-
TimeRewarder: Learning Dense Reward from Passive Videos via Frame-wise Temporal Distance
TimeRewarder derives progress-based dense rewards from passive videos via frame-wise temporal distance modeling and uses them as proxy rewards to boost RL success on Meta-World tasks.
-
LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning
LLM-TALE steers RL exploration using LLM-generated plans at task and affordance levels with online suboptimality correction, improving sample efficiency and success rates on pick-and-place tasks without human supervision.
-
Reinforcement Learning with Foundation Priors: Let the Embodied Agent Efficiently Learn on Its Own
RLFP and the FAC algorithm combine foundation-model priors for policy, value, and rewards to produce sample-efficient robotic RL that reaches 86% real-robot success after one hour and 100% success on 7/8 Meta-world ta...
-
Efficient On-policy Visual-RL via Stochastic Decoupled Policy Gradient
SDPG is a new on-policy visual RL algorithm that estimates gradients via stochastic perturbations of rollouts, achieving faster training and lower memory use than baselines on visual MuJoCo tasks while adding new robo...
-
Balancing Plasticity and Stability with Fast and Slow Successor Features
Synaptic consolidation applied to multi-timescale successor features yields better performance than plasticity-focused methods in RL under gradual environmental drift.
-
Tactile-based Multimodal Fusion in Embodied Intelligence: A Survey of Vision, Language, and Contact-Driven Paradigms
A survey proposing a hierarchical taxonomy for multimodal tactile fusion datasets and methods across perception, generation, and interaction in embodied intelligence.
Discussion (0). Sign in to comment.