Zobrazeno 1 - 10
of 14
pro vyhledávání: '"Jiang, Heinrich"'
Detecting out-of-distribution (OOD) examples is critical in many applications. We propose an unsupervised method to detect OOD samples using a $k$-NN density estimate with respect to a classification model's intermediate activations on in-distributio
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7e01fed0dd47179e43049d5b791a79bc
Self-supervised contrastive representation learning has proved incredibly successful in the vision and natural language domains, enabling state-of-the-art performance with orders of magnitude less labeled data. However, such methods are domain-specif
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::af6ae266c760271e6565bfa83e14fb8f
Autor:
Jiang, Heinrich, Rostamizadeh, Afshin
We analyze the problem of active covering, where the learner is given an unlabeled dataset and can sequentially label query examples. The objective is to label query all of the positive examples in the fewest number of total label queries. We show un
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::56ecf27aaea68ab82d8a62441e5101b7
We propose an approach to fair classification that enforces independence between the classifier outputs and sensitive information by minimizing Wasserstein-1 distances. The approach has desirable theoretical properties and is robust to specific choic
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5106cce11deac5444366dadbca214385
Autor:
Jiang, Heinrich, Nachum, Ofir
Datasets often contain biases which unfairly disadvantage certain groups, and classifiers trained on such datasets can inherit these biases. In this paper, we provide a mathematical formulation of how this bias can arise. We do so by assuming the exi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e20d839fc707f166677cc05355ca7e4c
Autor:
Nachum, Ofir, Jiang, Heinrich
A number of machine learning (ML) methods have been proposed recently to maximize model predictive accuracy while enforcing notions of group parity or fairness across sub-populations. We propose a desirable property for these procedures, slack-consis
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c4204d33657e925aaa4e31f48671b24d
Knowing when a classifier's prediction can be trusted is useful in many applications and critical for safely using AI. While the bulk of the effort in machine learning research has been towards improving classifier performance, understanding when a c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bcdc9078f2aa658a500fcc1508244cea
Autor:
Cotter, Andrew, Gupta, Maya, Jiang, Heinrich, Muller, James, Narayan, Taman, Wang, Serena, Zhu, Tao
We propose learning flexible but interpretable functions that aggregate a variable-length set of permutation-invariant feature vectors to predict a label. We use a deep lattice network model so we can architect the model structure to enhance interpre
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e456a19029e21f7a05762c4c4f5ca1fe
Autor:
Jang, Jennifer, Jiang, Heinrich
DBSCAN is a classical density-based clustering procedure with tremendous practical relevance. However, DBSCAN implicitly needs to compute the empirical density for each sample point, leading to a quadratic worst-case time complexity, which is too slo
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fe4e3b5eb2d3a989834584175df5bac7
Autor:
Cotter, Andrew, Jiang, Heinrich, Wang, Serena, Narayan, Taman, Gupta, Maya, You, Seungil, Sridharan, Karthik
We show that many machine learning goals, such as improved fairness metrics, can be expressed as constraints on the model's predictions, which we call rate constraints. We study the problem of training non-convex models subject to these rate constrai
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9bcbe7a09232059a4e2546971ed9b89c