Related papers: Doing More With Less: Mismatch-Based Risk-Limiting…
AI coding scaffolds like Claude Code and Codex use retrying: blocking actions flagged as risky and continuing the trajectory. We study retrying from an AI control perspective, which treats the model as potentially adversarial. We find that…
This paper provides a statistical analysis of high-dimensional batch Reinforcement Learning (RL) using sparse linear function approximation. When there is a large number of candidate features, our result sheds light on the fact that…
In safety-critical decision-making, the environment may evolve over time, and the learner adjusts its risk level accordingly. This work investigates risk-averse online optimization in dynamic environments with varying risk levels, employing…
In reinforcement learning, specifying reward functions that capture the intended task can be very challenging. Reward learning aims to address this issue by learning the reward function. However, a learned reward model may have a low error…
Misalignment in Large Language Models (LLMs) arises when model behavior diverges from human expectations and fails to simultaneously satisfy safety, value, and cultural dimensions, which must co-occur in real-world settings to solve a…
Robust estimators for linear regression require non-convex objective functions to shield against adverse affects of outliers. This non-convexity brings challenges, particularly when combined with penalization in high-dimensional settings.…
Randomized controlled trials (RCTs) are the gold standard for evaluating the causal effect of a treatment; however, they often have limited sample sizes and sometimes poor generalizability. On the other hand, non-randomized, observational…
Machine learning (ML) models often exhibit bias that can exacerbate inequities in biomedical applications. Fairness auditing, the process of evaluating a model's performance across subpopulations, is critical for identifying and mitigating…
While randomized controlled trials (RCTs) are the gold standard for estimating treatment effects in medical research, there is increasing use of and interest in using real-world data for drug development. One such use case is the…
Cross-Context Review (CCR) improves LLM verification by separating production and review into independent sessions. A natural extension is multi-turn review: letting the reviewer ask follow-up questions, receive author responses, and review…
We tackle the problem of estimating risk measures of the infinite-horizon discounted cost within a Markov cost process. The risk measures we study include variance, Value-at-Risk (VaR), and Conditional Value-at-Risk (CVaR). First, we show…
High-quality data annotation is an essential but laborious and costly aspect of developing machine learning-based software. We explore the inherent tradeoff between annotation accuracy and cost by detecting and removing minority reports --…
Algorithmic risk assessment instruments (RAIs) increasingly inform decision-making in criminal justice. RAIs largely rely on arrest records as a proxy for underlying crime. Problematically, the extent to which arrests reflect overall…
In this paper we study a family of variance reduction methods with randomized batch size---at each step, the algorithm first randomly chooses the batch size and then selects a batch of samples to conduct a variance-reduced stochastic…
The entropic value-at-risk (EVaR) is a new coherent risk measure, which is an upper bound for both the value-at-risk (VaR) and conditional value-at-risk (CVaR). As important properties, the EVaR is strongly monotone over its domain and…
Multi-view spectral clustering, which aims at yielding an agreement or consensus data objects grouping across multi-views with their graph laplacian matrices, is a fundamental clustering problem. Among the existing methods, Low-Rank…
We study a risk-constrained version of the stochastic shortest path (SSP) problem, where the risk measure considered is Conditional Value-at-Risk (CVaR). We propose two algorithms that obtain a locally risk-optimal policy by employing four…
Algorithmic lending has transformed the consumer credit landscape, with complex machine learning models now commonly used to make or assist underwriting decisions. To comply with fair lending laws, these algorithms typically exclude legally…
We study multiwinner elections with approval-based preferences. An instance of a multiwinner election consists of a set of alternatives, a population of voters---each voter approves a subset of alternatives, and the desired committee size…
We develop several tools for the determination of sample size and design for Medicaid and healthcare audits. The goal of these audits is to examine a population of claims submitted by a healthcare provider for reimbursement by a third party…