Learning to walk via deep reinforcement learning
Deep reinforcement learning (deep RL) holds the promise of automating the acquisition of
complex controllers that can map sensory inputs directly to low-level actions. In the domain …
complex controllers that can map sensory inputs directly to low-level actions. In the domain …
Re-examining linear embeddings for high-dimensional Bayesian optimization
B Letham, R Calandra, A Rai… - Advances in neural …, 2020 - proceedings.neurips.cc
Bayesian optimization (BO) is a popular approach to optimize expensive-to-evaluate black-
box functions. A significant challenge in BO is to scale to high-dimensional parameter …
box functions. A significant challenge in BO is to scale to high-dimensional parameter …
Increasing the scope as you learn: Adaptive Bayesian optimization in nested subspaces
L Papenmeier, L Nardi… - Advances in Neural …, 2022 - proceedings.neurips.cc
Recent advances have extended the scope of Bayesian optimization (BO) to expensive-to-
evaluate black-box functions with dozens of dimensions, aspiring to unlock impactful …
evaluate black-box functions with dozens of dimensions, aspiring to unlock impactful …
Learning fast adaptation with meta strategy optimization
The ability to walk in new scenarios is a key milestone on the path toward real-world
applications of legged robots. In this work, we introduce Meta Strategy Optimization, a …
applications of legged robots. In this work, we introduce Meta Strategy Optimization, a …
DeepWalk: Omnidirectional bipedal gait by deep reinforcement learning
D Rodriguez, S Behnke - 2021 IEEE international conference …, 2021 - ieeexplore.ieee.org
Bipedal walking is one of the most difficult but exciting challenges in robotics. The difficulties
arise from the complexity of high-dimensional dynamics, sensing and actuation limitations …
arise from the complexity of high-dimensional dynamics, sensing and actuation limitations …
Rethinking optimization with differentiable simulation from a global perspective
R Antonova, J Yang… - Conference on Robot …, 2023 - proceedings.mlr.press
Differentiable simulation is a promising toolkit for fast gradient-based policy optimization and
system identification. However, existing approaches to differentiable simulation have largely …
system identification. However, existing approaches to differentiable simulation have largely …
Geometry-aware Bayesian optimization in robotics using Riemannian Matérn kernels
Bayesian optimization is a data-efficient technique which can be used for control parameter
tuning, parametric policy adaptation, and structure design in robotics. Many of these …
tuning, parametric policy adaptation, and structure design in robotics. Many of these …
Bayesian optimization meets Riemannian manifolds in robot learning
Bayesian optimization (BO) recently became popular in robotics to optimize control
parameters and parametric policies in direct reinforcement learning due to its data efficiency …
parameters and parametric policies in direct reinforcement learning due to its data efficiency …
Using deep reinforcement learning to learn high-level policies on the atrias biped
Learning controllers for bipedal robots is a challenging problem, often requiring expert
knowledge and extensive tuning of parameters that vary in different situations. Recently …
knowledge and extensive tuning of parameters that vary in different situations. Recently …
[HTML][HTML] Learning to use chopsticks in diverse gripping styles
Learning dexterous manipulation skills is a long-standing challenge in computer graphics
and robotics, especially when the task involves complex and delicate interactions between …
and robotics, especially when the task involves complex and delicate interactions between …