133 citations · 282 across the 33 of their papers we have counts for
7 papers · 1 filter
Anytime-Valid Continuous-Time Confidence Processes for Inhomogeneous Poisson Processes
Michael Lindon, Nathan Kallus
Motivated by monitoring the arrival of incoming adverse events such as customer support calls or crash reports from users exposed to an experimental product change, we consider seq…
CSPI-MT: Calibrated Safe Policy Improvement with Multiple Testing for Threshold Policies
Brian M Cho, Ana-Roxana Pop, Kyra Gan +4
When modifying existing policies in high-risk settings, it is often necessary to ensure with high certainty that the newly proposed policy improves upon a baseline, such as the sta…
Nonparametric Jackknife Instrumental Variable Estimation and Confounding Robust Surrogate Indices
Aurélien Bibaut, Nathan Kallus, Apoorva Lal
Jackknife instrumental variable estimation (JIVE) is a classic method to leverage many weak instrumental variables (IVs) to estimate linear structural models, overcoming the bias o…
Reindex-Then-Adapt: Improving Large Language Models for Conversational Recommendation
Zhankui He, Zhouhang Xie, Harald Steck +4
Large language models (LLMs) are revolutionizing conversational recommender systems by adeptly indexing item content, understanding complex conversational contexts, and generating…
Demistifying Inference after Adaptive Experiments
Aurélien Bibaut, Nathan Kallus
Adaptive experiments such as multi-arm bandits adapt the treatment-allocation policy and/or the decision to stop the experiment to the data observed so far. This has the potential…
Is Cosine-Similarity of Embeddings Really About Similarity?
Harald Steck, Chaitanya Ekanadham, Nathan Kallus
Cosine-similarity is the cosine of the angle between two vectors, or equivalently the dot product between their normalizations. A popular application is to quantify semantic simila…
Learning the Covariance of Treatment Effects Across Many Weak Experiments
Aurélien Bibaut, Winston Chou, Simon Ejdemyr +1
When primary objectives are insensitive or delayed, experimenters may instead focus on proxy metrics derived from secondary outcomes. For example, technology companies often infer…