• Publications
  • Influence
Extreme F-measure Maximization using Sparse Probability Estimates
TLDR
This work considers the problem of (macro) F-measure maximization in the context of extreme multilabel classification (XMLC) and proposes to solve the problem by classifiers that efficiently deliver sparse probability estimates (SPEs), that is, probability estimates restricted to the most probable labels.
Online Rank Elicitation for Plackett-Luce: A Dueling Bandits Approach
TLDR
The approach is based on constructing a surrogate probability distribution over rankings based on a sorting procedure, for which the pairwise marginals provably coincide with the marginals of the Plackett-Luce distribution.
Top-k Selection based on Adaptive Sampling of Noisy Preferences
TLDR
This work proposes and formally analyze a general preference-based racing algorithm that is instantiate with three specific ranking procedures and corresponding sampling schemes, and assumes that alternatives can be compared in terms of pairwise preferences.
Qualitative Multi-Armed Bandits: A Quantile-Based Approach
TLDR
This work formalizes and study the multi-armed bandit problem in a generalized stochastic setting, in which rewards are not assumed to be numerical, and addresses the problem of quantile-based online learning both for the case of a finite and infinite time horizon.
Boosting products of base classifiers
TLDR
This paper shows how to boost products of simple base learners and presents an improved base learner for nominal features and shows that boosting the product of two of these new subset indicator base learners solves the maximum margin matrix factorization problem used to formalize the collaborative filtering task.
A no-regret generalization of hierarchical softmax to extreme multi-label classification
TLDR
It is shown that PLTs are a no-regret multi-label generalization of HSM when precision@$k$ is used as a model evaluation metric, and it is proved that pick-one-label heuristic---a reduction technique from multi- label to multi-class that is routinely used along with HSM---is not consistent in general.
Bacterial evolution of antibiotic hypersensitivity
TLDR
Using large‐scale laboratory evolutionary experiments with Escherichia coli, it is demonstrated that collateral sensitivity occurs frequently during the evolution of antibiotic resistance, and insight is offered into the mechanisms that drive the Evolution of negative trade‐offs under antibiotic selection.
Online F-Measure Optimization
TLDR
This paper proposes an efficient online algorithm for F-measure maximization and provides a formal analysis of its convergence properties and first experimental results are presented, showing that the method performs well in practice.
State-of-the-art anonymization of medical records using an iterative machine learning framework.
TLDR
A de-identification model that can successfully remove personal health information (PHI) from discharge records to make them conform to the guidelines of the Health Information Portability and Accountability Act is developed.
Fast boosting using adversarial bandits
TLDR
This paper applies multi-armed bandits (MABs) to improve the computational complexity of AdaBoost and uses an adversarial bandit algorithm instead of stochastic bandits to prove a weak-to-strong-learning theorem.
...
1
2
3
4
5
...