Safe Exploration

30 papers with code • 0 benchmarks • 0 datasets

Safe Exploration is an approach to collect ground truth data by safely interacting with the environment.

Source: Chance-Constrained Trajectory Optimization for Safe Exploration and Learning of Nonlinear Systems


Use these libraries to find Safe Exploration models and implementations
2 papers

Most implemented papers

Safe Exploration in Continuous Action Spaces

utiasDSL/safe-control-gym 26 Jan 2018

We address the problem of deploying a reinforcement learning (RL) agent on a physical system such as a datacenter cooling unit or robot, where critical constraints must never be violated.

Feasible Actor-Critic: Constrained Reinforcement Learning for Ensuring Statewise Safety

mahaitongdae/Feasible-Actor-Critic 22 May 2021

The safety constraints commonly used by existing safe reinforcement learning (RL) methods are defined only on expectation of initial states, but allow each certain state to be unsafe, which is unsatisfying for real-world safety-critical tasks.

AI Safety Gridworlds

deepmind/ai-safety-gridworlds 27 Nov 2017

We present a suite of reinforcement learning environments illustrating various safety properties of intelligent agents.

SafeML: Safety Monitoring of Machine Learning Classifiers through Statistical Difference Measure

ISorokos/SafeML 27 May 2020

Ensuring safety and explainability of machine learning (ML) is a topic of increasing relevance as data-driven applications venture into safety-critical application domains, traditionally committed to high safety standards that are not satisfied with an exclusive testing approach of otherwise inaccessible black-box systems.

MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning

decisionforce/metadrive 26 Sep 2021

Based on MetaDrive, we construct a variety of RL tasks and baselines in both single-agent and multi-agent settings, including benchmarking generalizability across unseen scenes, safe exploration, and learning multi-agent traffic.

Safe Exploration in Finite Markov Decision Processes with Gaussian Processes

befelix/SafeMDP NeurIPS 2016

We define safety in terms of an, a priori unknown, safety constraint that depends on states and actions.

Concrete Problems in AI Safety

mateuszjurewicz/bornhack_ml_crashcourse 21 Jun 2016

Rapid progress in machine learning and artificial intelligence (AI) has brought increasing attention to the potential impacts of AI technologies on society.

Learning-based Model Predictive Control for Safe Exploration

befelix/safe-exploration 22 Mar 2018

However, these methods typically do not provide any safety guarantees, which prevents their use in safety-critical, real-world applications.

Learning-based Model Predictive Control for Safe Exploration and Reinforcement Learning

befelix/safe-exploration 27 Jun 2019

We evaluate the resulting algorithm to safely explore the dynamics of an inverted pendulum and to solve a reinforcement learning task on a cart-pole system with safety constraints.

Safe Exploration for Optimizing Contextual Bandits

rjagerman/tois2019-safe-exploration-algorithm 2 Feb 2020

Our experiments using text classification and document retrieval confirm the above by comparing SEA (and a boundless variant called BSEA) to online and offline learning methods for contextual bandit problems.