English
Related papers

Related papers: Defining and Characterizing Reward Hacking

200 papers

The paper addresses the problem of computing maximal conditional expected accumulated rewards until reaching a target state (briefly called maximal conditional expectations) in finite-state Markov decision processes where the condition is…

Logic in Computer Science · Computer Science 2023-03-07 Christel Baier , Joachim Klein , Sascha Klüppelholz , Sascha Wunderlich

Empirical researchers and decision-makers spanning various domains frequently seek profound insights into the long-term impacts of interventions. While the significance of long-term outcomes is undeniable, an overemphasis on them may…

Machine Learning · Computer Science 2024-09-17 Peng Wu , Ziyu Shen , Feng Xie , Zhongyao Wang , Chunchen Liu , Yan Zeng

Variant Stochastic cracking is a significantly more resilient approach to adaptive indexing. It showed [1]that Stochastic cracking uses each query as a hint on how to reorganize data, but not blindly so; it gains resilience and avoids…

Databases · Computer Science 2013-05-09 Meenesh Bhardwaj

Would it be possible to explain the emergence of new computational ideas using the computation itself? Would it be feasible to describe the discovery process of new algorithmic solutions using only mathematics? This study is the first…

Selective rationalization has become a common mechanism to ensure that predictive models reveal how they use any available features. The selection may be soft or hard, and identifies a subset of input features relevant for prediction. The…

Computation and Language · Computer Science 2019-12-17 Mo Yu , Shiyu Chang , Yang Zhang , Tommi S. Jaakkola

Designing reward functions is difficult: the designer has to specify what to do (what it means to complete the task) as well as what not to do (side effects that should be avoided while completing the task). To alleviate the burden on the…

Machine Learning · Computer Science 2020-10-16 Victoria Krakovna , Laurent Orseau , Richard Ngo , Miljan Martic , Shane Legg

When designing optimal controllers for any system, it is often the case that the true state of the system is unknown to the controller, for example due to noisy measurements or partially observable states. Incomplete state information must…

Optimization and Control · Mathematics 2014-01-23 Kendra Lesser , Meeko Oishi

In many reinforcement learning (RL) applications, augmenting the task rewards with heuristic rewards that encode human priors about how a task should be solved is crucial for achieving desirable performance. However, because such heuristics…

Machine Learning · Computer Science 2025-07-09 Chi-Chang Lee , Zhang-Wei Hong , Pulkit Agrawal

In many real-world tasks, it is not possible to procedurally specify an RL agent's reward function. In such cases, a reward function must instead be learned from interacting with and observing humans. However, current techniques for reward…

Machine Learning · Computer Science 2020-12-11 Eric J. Michaud , Adam Gleave , Stuart Russell

Agent benchmarks have become the de facto measure of frontier AI competence, guiding model selection, investment, and deployment. However, reward hacking, where agents maximize a score without performing the intended task, emerges…

Artificial Intelligence · Computer Science 2026-05-14 Hao Wang , Hanchen Li , Qiuyang Mang , Alvin Cheung , Koushik Sen , Dawn Song

In the standard setting of approachability there are two players and a target set. The players play repeatedly a known vector-valued game where the first player wants to have the average vector-valued payoff converge to the target set which…

Machine Learning · Statistics 2016-06-20 Shie Mannor , Vianney Perchet , Gilles Stoltz

Language models trained with reinforcement learning (RL) can engage in reward hacking--the exploitation of unintended strategies for high reward--without revealing this behavior in their chain-of-thought reasoning. This makes the detection…

Computation and Language · Computer Science 2025-07-15 Miles Turpin , Andy Arditi , Marvin Li , Joe Benton , Julian Michael

We initiate the study of incentive-compatible forecasting competitions in which multiple forecasters make predictions about one or more events and compete for a single prize. We have two objectives: (1) to incentivize forecasters to report…

Computer Science and Game Theory · Computer Science 2021-09-09 Jens Witkowski , Rupert Freeman , Jennifer Wortman Vaughan , David M. Pennock , Andreas Krause

Algorithmic fairness involves expressing notions such as equity, or reasonable treatment, as quantifiable measures that a machine learning algorithm can optimise. Most work in the literature to date has focused on classification problems…

Machine Learning · Computer Science 2020-03-06 Daniel Steinberg , Alistair Reid , Simon O'Callaghan

The intersection of causal inference and machine learning for decision-making is rapidly expanding, but the default decision criterion remains an \textit{average} of individual causal outcomes across a population. In practice, various…

Machine Learning · Computer Science 2022-11-08 Wenshuo Guo , Michael I. Jordan , Angela Zhou

Terminal-agent benchmarks have become a primary signal for measuring the coding and system-administration capabilities of large language models. As the market for evaluation environments grows, so does the pressure to ship tasks quickly,…

Artificial Intelligence · Computer Science 2026-05-01 Ivan Bercovich

Recent advances in aligning Large Language Models with human preferences have benefited from larger reward models and better preference data. However, most of these methodologies rely on the accuracy of the reward model. The reward models…

Artificial Intelligence · Computer Science 2024-11-01 Debangshu Banerjee , Aditya Gopalan

Explainable AI (XAI) and interpretable machine learning methods help to build trust in model predictions and derived insights, yet also present a perverse incentive for analysts to manipulate XAI metrics to support pre-specified…

Machine Learning · Computer Science 2025-07-16 Rahul Sharma , Sergey Redyuk , Sumantrak Mukherjee , Andrea Šipka , Eyke Hüllermeier , Sebastian Vollmer , David Selby

Opacity is a generic security property, that has been defined on (non probabilistic) transition systems and later on Markov chains with labels. For a secret predicate, given as a subset of runs, and a function describing the view of an…

Cryptography and Security · Computer Science 2014-09-02 Béatrice Bérard , Krishnendu Chatterjee , Nathalie Sznajder

Reinforcement learning (RL) methods usually treat reward functions as black boxes. As such, these methods must extensively interact with the environment in order to discover rewards and optimal policies. In most RL applications, however,…

Machine Learning · Computer Science 2022-01-19 Rodrigo Toro Icarte , Toryn Q. Klassen , Richard Valenzano , Sheila A. McIlraith
‹ Prev 1 8 9 10 Next ›