paper-with-me

Papers

DropoutDAgger: A Bayesian Approach to Safe Imitation Learning

2017-09-18 · Kunal Menda, Katherine Driggs-Campbell, Mykel J. Kochenderfer

While imitation learning is becoming common practice in robotics, this approach often suffers from data mismatch and compounding errors. DAgger is an iterative algorithm that addresses these issues by continually aggregating training data from both the expert and novice policies, but does not consider the impact of safety. We present a probabilistic extension to DAgger, which uses the distribution over actions provided by the novice policy, for a given observation. Our method, which we call DropoutDAgger, uses dropout to train the novice as a Bayesian neural network that provides insight to its confidence. Using the distribution over the novice's actions, we estimate a probabilistic measure of safety with respect to the expert action, tuned to balance exploration and exploitation. The utility of this approach is evaluated on the MuJoCo HalfCheetah and in a simple driving experiment, demonstrating improved performance and safety compared to other DAgger variants and classic imitation learning.

📄 PDF Abstract BibTeX arXiv:1709.06166

Code (0)

등록된 구현이 없습니다.

Tasks

Imitation LearningMuJoCo

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

Safe Imitation Learning via Fast Bayesian Reward Inference from Preferences

2020-02-21 · ICML 2020 1 · Daniel S. Brown, Russell Coleman, Ravi Srinivasan, Scott Niekum

Bayesian reward learning from demonstrations enables rigorous safety and uncertainty analysis when performing imitation learning. However, Bayesian reward learning methods are typically computationally intractable for co…

Atari GamesBayesian InferenceImitation Learning

Deep Bayesian Reward Learning from Preferences

2019-12-10 · Daniel S. Brown, Scott Niekum

Bayesian inverse reinforcement learning (IRL) methods are ideal for safe imitation learning, as they allow a learning agent to reason about reward uncertainty and the safety of a learned policy. However, Bayesian IRL is …

Atari GamesImitation LearningReinforcement Learning

Tuning Particle Accelerators with Safety Constraints using Bayesian Optimization

2022-03-26 · Johannes Kirschner, Mojmir Mutný, Andreas Krause, Jaime Coello de Portugal 외

Tuning machine parameters of particle accelerators is a repetitive and time-consuming task that is challenging to automate. While many off-the-shelf optimization algorithms are available, in practice their use is limited…

Bayesian Optimizationglobal-optimization

Safety-Aware Robot Damage Recovery Using Constrained Bayesian Optimization and Simulated Priors

2016-11-28 · Vaios Papaspyros, Konstantinos Chatzilygeroudis, Vassilis Vassiliades, Jean-Baptiste Mouret

The recently introduced Intelligent Trial-and-Error (IT&E) algorithm showed that robots can adapt to damage in a matter of a few trials. The success of this algorithm relies on two components: prior knowledge acquired th…

Bayesian Optimization

Bayesian Deep Learning for Segmentation for Autonomous Safe Planetary Landing

2021-02-21 · Kento Tomita, Katherine A. Skinner, Koki Ho

Hazard detection is critical for enabling autonomous landing on planetary surfaces. Current state-of-the-art methods leverage traditional computer vision approaches to automate the identification of safe terrain from inp…

Deep LearningSemantic Segmentation