Safe Exploration Techniques for Reinforcement Learning – An Overview

Pecka, Martin; Svoboda, Tomáš

doi:10.1007/978-3-319-13823-7_31

Cited by 56 publications

(38 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, tailoring the insulin delivery policy of an artificial pancreas to the metabolism of an individual requires trial insulin delivery action but these should only be sampled when their outcome is within safe certainty bounds [44]. If safety is a significant concern in the systems' application domain, specifically designed safety-aware RL techniques may be required, see [149] and [64] for overviews of such techniques.…”

Section: A Classification Of Personalization Settingsmentioning

confidence: 99%

Reinforcement learning for personalization: A systematic literature review

Hengst

Grua

Hassouni

et al. 2020

View full text Add to dashboard Cite

The major application areas of reinforcement learning (RL) have traditionally been game playing and continuous control. In recent years, however, RL has been increasingly applied in systems that interact with humans. RL can personalize digital systems to make them more relevant to individual users. Challenges in personalization settings may be different from challenges found in traditional application areas of RL. An overview of work that uses RL for personalization, however, is lacking. In this work, we introduce a framework of personalization settings and use it in a systematic literature review. Besides setting, we review solutions and evaluation strategies. Results show that RL has been increasingly applied to personalization problems and realistic evaluations have become more prevalent. RL has become sufficiently robust to apply in contexts that involve humans and the field as a whole is growing. However, it seems not to be maturing: the ratios of studies that include a comparison or a realistic evaluation are not showing upward trends and the vast majority of algorithms are used only once. This review can be used to find related work across domains, provides insights into the state of the field and identifies opportunities for future work.

show abstract

Section: A Classification Of Personalization Settingsmentioning

confidence: 99%

Reinforcement learning for personalization: A systematic literature review

Hengst

Grua

Hassouni

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Some directly enforce a set of safety rules to prevent certain actions leading to unsafe states (90); some encode task safety specifications in the reward function to encourage agents to explore safely (91). An overview for safe exploration in reinforcement learning can be found in (92). Although safe exploration reduces exploration failures, which helps maintain exploration continuity, it may limit exploration to a subspace so that the search may not be thorough (93).…”

Section: Active Learning and Explorationmentioning

confidence: 99%

Toward next-generation learned robot manipulation

Cui

Trinkle

2021

Sci. Robot.

View full text Add to dashboard Cite

The ever-changing nature of human environments presents great challenges to robot manipulation. Objects that robots must manipulate vary in shape, weight, and configuration. Important properties of the robot, such as surface friction and motor torque constants, also vary over time. Before robot manipulators can work gracefully in homes and businesses, they must be adaptive to such variations. This survey summarizes types of variations that robots may encounter in human environments and categorizes, compares, and contrasts the ways in which learning has been applied to manipulation problems through the lens of adaptability. Promising avenues for future research are proposed at the end.

show abstract

“…In this work, we conceive a dynamic pandemic lockdown strategy that factors in public health infrastructure of a geographical region. The proposed approach built upon reinforcement learning (RL) allows agents to take decisions to maximize reward, while adapting to a complex and uncertain environment (Tuyls and Weiss 2012;Pecka and Svoboda 2014). We create an agent-based simulation environment running the ordinary differential equation-based SEIRD epidemic model (Hethcote 2000) (discussed in "Scenario" section).…”

Section: Issues In Vaccine Production and Supplymentioning

confidence: 99%

Towards Dynamic lockdown strategies controlling pandemic spread under healthcare resource budget

2021

View full text Add to dashboard Cite

COVID-19 is one of the deadliest pandemics in modern human history that has killed nearly a million people and rapidly inundated the healthcare resources around the world. Current lockdown measures to curb infection spread are threatening to bring the world economy to a halt, necessitating dynamic lockdown policies that incorporate the healthcare resource budget of people in a zone. We conceive a dynamic pandemic lockdown strategy that employs reinforcement learning to modulate the zone mobility, while restricting the COVID-19 hospitalizations within its healthcare resource budget. We employ queueing theory to model the inflow and outflow of patients and validate the approach through extensive simulation on real demographic and epidemiological data from the boroughs of New York City. Our experiments demonstrate that this approach can not only adapt to the varying trends in contagion in a region by regulating its own lockdown level, but also manages the overheads associated with time-varying dynamic lockdown policies.

show abstract

Safe Exploration Techniques for Reinforcement Learning – An Overview

Cited by 56 publications

References 16 publications

Reinforcement learning for personalization: A systematic literature review

Reinforcement learning for personalization: A systematic literature review

Toward next-generation learned robot manipulation

Towards Dynamic lockdown strategies controlling pandemic spread under healthcare resource budget

Contact Info

Product

Resources

About