DDPG is a model-free actor-critic algorithm that learns continuous control policies end-to-end from states or pixels using deterministic policy gradients and deep networks, solving more than 20 physics tasks competitively with full-information planning methods.
Memory-based control with recurrent neural networks
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
Partially observed control problems are a challenging aspect of reinforcement learning. We extend two related, model-free algorithms for continuous control -- deterministic policy gradient and stochastic value gradient -- to solve partially observed domains using recurrent neural networks trained with backpropagation through time. We demonstrate that this approach, coupled with long-short term memory is able to solve a variety of physical control problems exhibiting an assortment of memory requirements. These include the short-term integration of information from noisy sensors and the identification of system parameters, as well as long-term memory problems that require preserving information over many time steps. We also demonstrate success on a combined exploration and memory problem in the form of a simplified version of the well-known Morris water maze task. Finally, we show that our approach can deal with high-dimensional observations by learning directly from pixels. We find that recurrent deterministic and stochastic policies are able to learn similarly good solutions to these tasks, including the water maze where the agent must learn effective search strategies.
representative citing papers
Compressive Transformer sets new records on WikiText-103 (17.1 ppl) and Enwik8 (0.97 bpc) via memory compression and introduces the PG-19 long-range language benchmark.
A decoupled estimator combining gated dynamics learning and recursive Kalman filtering improves robustness of pre-trained MARL policies under stale observations and message loss.
The authors compare multiple methods for incorporating action information into RNN state updates for RL and report empirical results on illustrative domains.
Offline RL for ICU sedation shows that adding 30-day mortality to the objective yields policies whose clinician agreement correlates negatively with mortality, unlike pain-only versions.
L-Learning learns the system's energy function from data to deliver stable, accurate, and sample-efficient robot trajectory tracking with closed-loop stability guarantees.
citing papers explorer
-
L-Learning : A Lyapunov-Based Approach Leveraging Lagrangian Mechanics for Efficient and Stable Robot Tracking
L-Learning learns the system's energy function from data to deliver stable, accurate, and sample-efficient robot trajectory tracking with closed-loop stability guarantees.