Intelligent problem-solving as integrated hierarchical reinforcement learning
According to cognitive psychology and related disciplines, the development of complex
problem-solving behaviour in biological agents depends on hierarchical cognitive …
problem-solving behaviour in biological agents depends on hierarchical cognitive …
Learning agile soccer skills for a bipedal robot with deep reinforcement learning
We investigated whether deep reinforcement learning (deep RL) is able to synthesize
sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be …
sophisticated and safe movement skills for a low-cost, miniature humanoid robot that can be …
From motor control to team play in simulated humanoid football
Learning to combine control at the level of joint torques with longer-term goal-directed
behavior is a long-standing challenge for physically embodied artificial agents. Intelligent …
behavior is a long-standing challenge for physically embodied artificial agents. Intelligent …
The challenges of exploration for offline reinforcement learning
Offline Reinforcement Learning (ORL) enablesus to separately study the two interlinked
processes of reinforcement learning: collecting informative experience and inferring optimal …
processes of reinforcement learning: collecting informative experience and inferring optimal …
Is bang-bang control all you need? solving continuous control with bernoulli policies
Reinforcement learning (RL) for continuous control typically employs distributions whose
support covers the entire action space. In this work, we investigate the colloquially known …
support covers the entire action space. In this work, we investigate the colloquially known …
Independent component alignment for multi-task learning
D Senushkin, N Patakin… - Proceedings of the …, 2023 - openaccess.thecvf.com
In a multi-task learning (MTL) setting, a single model is trained to tackle a diverse set of tasks
jointly. Despite rapid progress in the field, MTL remains challenging due to optimization …
jointly. Despite rapid progress in the field, MTL remains challenging due to optimization …
Data-efficient hindsight off-policy option learning
Abstract We introduce Hindsight Off-policy Options (HO2), a data-efficient option learning
algorithm. Given any trajectory, HO2 infers likely option choices and backpropagates …
algorithm. Given any trajectory, HO2 infers likely option choices and backpropagates …
Measuring interpretability of neural policies of robots with disentangled representation
The advancement of robots, particularly those functioning in complex human-centric
environments, relies on control solutions that are driven by machine learning …
environments, relies on control solutions that are driven by machine learning …
Collect & infer-a fresh look at data-efficient reinforcement learning
M Riedmiller, JT Springenberg… - … on Robot Learning, 2022 - proceedings.mlr.press
This position paper proposes a fresh look at Reinforcement Learning (RL) from the
perspective of data-efficiency. RL has gone through three major stages: pure on-line RL …
perspective of data-efficiency. RL has gone through three major stages: pure on-line RL …
Behavior priors for efficient reinforcement learning
As we deploy reinforcement learning agents to solve increasingly challenging problems,
methods that allow us to inject prior knowledge about the structure of the world and effective …
methods that allow us to inject prior knowledge about the structure of the world and effective …