English
Related papers

Related papers: Golden Handcuffs make safer AI agents

200 papers

Mean rewards of actions are often correlated. The form of these correlations may be complex and unknown a priori, such as the preferences of a user for recommended products and their categories. To maximize statistical efficiency, it is…

Machine Learning · Computer Science 2022-02-04 Joey Hong , Branislav Kveton , Sumeet Katariya , Manzil Zaheer , Mohammad Ghavamzadeh

Autonomous agents that act with each other on behalf of humans are becoming more common in many social domains, such as customer service, transportation, and health care. In such social situations greedy strategies can reduce the positive…

Multiagent Systems · Computer Science 2022-12-02 Jory Schossau , Bamshad Shirmohammadi , Arend Hintze

Offline reinforcement learning (RL) is crucial for real-world applications where exploration can be costly or unsafe. However, offline learned policies are often suboptimal, and further online fine-tuning is required. In this paper, we…

Machine Learning · Computer Science 2024-06-03 Hao Hu , Yiqin Yang , Jianing Ye , Chengjie Wu , Ziqing Mai , Yujing Hu , Tangjie Lv , Changjie Fan , Qianchuan Zhao , Chongjie Zhang

Reward functions are easy to misspecify; although designers can make corrections after observing mistakes, an agent pursuing a misspecified reward function can irreversibly change the state of its environment. If that change precludes…

Artificial Intelligence · Computer Science 2020-06-11 Alexander Matt Turner , Dylan Hadfield-Menell , Prasad Tadepalli

This paper considers the use of a simple posterior sampling algorithm to balance between exploration and exploitation when learning to optimize actions such as in multi-armed bandit problems. The algorithm, also known as Thompson Sampling,…

Machine Learning · Computer Science 2014-02-04 Daniel Russo , Benjamin Van Roy

The human intrinsic desire to pursue knowledge, also known as curiosity, is considered essential in the process of skill acquisition. With the aid of artificial curiosity, we could equip current techniques for control, such as Reinforcement…

Machine Learning · Computer Science 2022-02-24 Pietro Mazzaglia , Ozan Catal , Tim Verbelen , Bart Dhoedt

When deploying artificial agents in real-world environments where they interact with humans, it is crucial that their behavior is aligned with the values, social norms or other requirements of that environment. However, many environments…

Machine Learning · Computer Science 2023-05-05 Mattijs Baert , Pietro Mazzaglia , Sam Leroux , Pieter Simoens

Current reinforcement learning algorithms train an agent using forward-generated trajectories, which provide little guidance so that the agent can explore as much as possible. While realizing the value of reinforcement learning results from…

Artificial Intelligence · Computer Science 2023-09-06 KyungMin Ko

Reinforcement learning is commonly concerned with problems of maximizing accumulated rewards in Markov decision processes. Oftentimes, a certain goal state or a subset of the state space attain maximal reward. In such a case, the…

Artificial Intelligence · Computer Science 2024-08-23 Pavel Osinenko , Grigory Yaremenko , Georgiy Malaniya , Anton Bolychev , Alexander Gepperth

Can artificial agents learn to assist others in achieving their goals without knowing what those goals are? Generic reinforcement learning agents could be trained to behave altruistically towards others by rewarding them for altruistic…

Artificial Intelligence · Computer Science 2022-03-22 Tim Franzmeyer , Mateusz Malinowski , João F. Henriques

Auctions in which agents' payoffs are random variables have received increased attention in recent years. In particular, recent work in algorithmic mechanism design has produced mechanisms employing internal randomization, partly in…

Computer Science and Game Theory · Computer Science 2012-06-15 Shaddin Dughmi , Yuval Peres

Reinforcement Learning AI commonly uses reward/penalty signals that are objective and explicit in an environment -- e.g. game score, completion time, etc. -- in order to learn the optimal strategy for task performance. However, Human-AI…

Human-Computer Interaction · Computer Science 2017-09-15 Victor Shih , David C Jangraw , Paul Sajda , Sameer Saproo

In the classic multi-armed bandits problem, the goal is to have a policy for dynamically operating arms that each yield stochastic rewards with unknown means. The key metric of interest is regret, defined as the gap between the expected…

Optimization and Control · Mathematics 2010-11-23 Yi Gai , Bhaskar Krishnamachari , Rahul Jain

Learning robot controllers by minimizing a black-box objective cost using Bayesian optimization (BO) can be time-consuming and challenging. It is very often the case that some roll-outs result in failure behaviors, causing premature…

Machine Learning · Computer Science 2020-11-11 Alonso Marco , Dominik Baumann , Philipp Hennig , Sebastian Trimpe

We study the effect of persistence of engagement on learning in a stochastic multi-armed bandit setting. In advertising and recommendation systems, repetition effect includes a wear-in period, where the user's propensity to reward the…

Machine Learning · Computer Science 2020-06-19 Priyank Agrawal , Theja Tulabandhula

Motivated by modern applications, such as online advertisement and recommender systems, we study the top-$k$ extreme contextual bandits problem, where the total number of arms can be enormous, and the learner is allowed to select $k$ arms…

Machine Learning · Statistics 2021-02-17 Rajat Sen , Alexander Rakhlin , Lexing Ying , Rahul Kidambi , Dean Foster , Daniel Hill , Inderjit Dhillon

We consider a resource-aware variant of the classical multi-armed bandit problem: In each round, the learner selects an arm and determines a resource limit. It then observes a corresponding (random) reward, provided the (random) amount of…

Machine Learning · Computer Science 2022-10-18 Viktor Bengs , Eyke Hüllermeier

Incentive design is a popular framework for guiding agents' learning dynamics towards desired outcomes by providing additional payments beyond intrinsic rewards. However, most existing works focus on a finite, small set of agents or assume…

Machine Learning · Computer Science 2025-04-16 Leo Widmer , Jiawei Huang , Niao He

Classic no-regret multi-armed bandit algorithms, including the Upper Confidence Bound (UCB), Hedge, and EXP3, are inherently unfair by design. Their unfairness stems from their objective of playing the most rewarding arm as frequently as…

Machine Learning · Computer Science 2024-05-14 Abhishek Sinha

As more machine learning agents interact with humans, it is increasingly a prospect that an agent trained to perform a task optimally, using only a measure of task performance as feedback, can violate societal norms for acceptable behavior…

Machine Learning · Computer Science 2021-04-20 Md Sultan Al Nahian , Spencer Frazier , Brent Harrison , Mark Riedl