Risk-averse autonomous systems: A brief history and recent developments from the perspective of optimal control

Y Wang, MP Chapman - Artificial Intelligence, 2022 - Elsevier
We present an historical overview about the connections between the analysis of risk and
the control of autonomous systems. We offer two main contributions. Our first contribution is …

On optimizing the conditional value-at-risk of a maximum cost for risk-averse safety analysis

MP Chapman, M Fauß, KM Smith - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
The popularity of Conditional Value-at-Risk (CVaR), a risk functional from finance, has been
growing in the control systems community due to its intuitive interpretation and axiomatic …

Stationary Policies are Optimal in Risk-averse Total-reward MDPs with EVaR

X Su, M Petrik, J Grand-Clément - arXiv preprint arXiv:2408.17286, 2024 - arxiv.org
Optimizing risk-averse objectives in discounted MDPs is challenging because most models
do not admit direct dynamic programming equations and require complex history-dependent …

[HTML][HTML] AoI-aware transmission control in real-time mmwave energy harvesting systems: a risk-sensitive reinforcement learning approach

M Sheikhi, V Hakami - Digital Communications and Networks, 2024 - Elsevier
The evolution of enabling technologies in wireless communications has paved the way for
supporting novel applications with more demanding QoS requirements, but at the cost of …

Optimality of Stationary Policies in Risk-averse Total-reward MDPs with EVaR

X Su, M Petrik, J Grand-Clément - ICML 2024 Workshop …, 2024 - openreview.net
The risk-neutral discounted objective is popular in reinforcement learning, in part due to
existence of stationary optimal policies and convenient analysis based on contracting …

A tighter problem-dependent regret bound for risk-sensitive reinforcement learning

X Hu, H Leung - International Conference on Artificial …, 2023 - proceedings.mlr.press
We study the regret for risk-sensitive reinforcement learning (RL) with the exponential utility
in the episodic MDP. Recent works establish both a lower bound $\Omega ((e^{|\beta|(H …

CVaR-based safety analysis in the infinite time horizon setting

C Wei, M Fauß, MP Chapman - 2022 American Control …, 2022 - ieeexplore.ieee.org
We develop a risk-averse safety analysis method for stochastic systems on discrete infinite
time horizons. Our method quantifies the notion of risk for a control system in terms of the …

Risk-aware stability of linear systems

MP Chapman, D Kalogerias - IEEE Transactions on Automatic …, 2024 - ieeexplore.ieee.org
We develop a generalized stability framework for stochastic discrete-time linear systems that
enriches the ways in which the distribution of the state energy can be characterized. We use …

Risk assessment of river water quality using long-memory processes subject to divergence or Wasserstein uncertainty

H Yoshioka, Y Yoshioka - Stochastic Environmental Research and Risk …, 2024 - Springer
River water quality often follows a long-memory stochastic process with power-type
autocorrelation decay, which can only be reproduced using appropriate mathematical …

Data-driven decision-making under uncertainty with entropic risk measure

U Sadana, E Delage, A Georghiou - arXiv preprint arXiv:2409.19926, 2024 - arxiv.org
The entropic risk measure is widely used in high-stakes decision making to account for tail
risks associated with an uncertain loss. With limited data, the empirical entropic risk …