California Institute of Technology
Author pages are created from data sourced from our academic publisher partnerships and public sources.
Share This Author
A support vector method for optimizing average precision
This work presents a general SVM learning algorithm that efficiently finds a globally optimal solution to a straightforward relaxation of MAP, and shows its method to produce statistically significant improvements in MAP scores.
Interactively optimizing information retrieval systems as a dueling bandits problem
An on-line learning framework tailored towards real-time learning from observed user behavior in search engines and other information retrieval systems and an algorithm with theoretical guarantees as well as simulation results are presented.
The K-armed Dueling Bandits Problem
Batch Policy Learning under Constraints
A new and simple method for off-policy policy evaluation (OPE) and derive PAC-style bounds is proposed and achieves strong empirical results in different domains, including in a challenging problem of simulated car driving subject to multiple constraints such as lane keeping and smooth driving.
Beat the Mean Bandit
This paper presents the first algorithm for this more general Dueling Bandits Problem and provides theoretical guarantees in both the online and the PAC settings and shows that the new algorithm has stronger guarantees than existing results even in the original DuelingBandits Problem, which is validated empirically.
Large-scale validation and analysis of interleaved search evaluation
This paper provides a comprehensive analysis of interleaving using data from two major commercial search engines and a retrieval system for scientific literature, and analyzes the agreement ofinterleaving with manual relevance judgments and observational implicit feedback measures.
Predicting diverse subsets using structural SVMs
This work formulate the learning problem of predicting diverse subsets and derive a training method based on structural SVMs that explicitly trains to diversify results.
Linear Submodular Bandits and their Application to Diversified Retrieval
This paper proposes the linear sub-modular bandits problem, which is an online learning setting for optimizing a general class of feature-rich submodular utility models for diversified retrieval, and presents an algorithm, called LSBGREEDY, and proves that it efficiently converges to a near-optimal model.
A deep learning approach for generalized speech animation
A simple and effective deep learning approach to automatically generate natural looking speech animation that synchronizes to input speech and can also generate on-demand speech animation in real-time from user speech input.
Beyond position bias: examining result attractiveness as a source of presentation bias in clickthrough data
This study distinguishes itself from prior work by aiming to detect systematic biases in click behavior due to attractive summaries inflating perceived relevance, and shows substantial evidence of presentation bias in clicks towards results with more attractive titles.