A Universal Unbiased Method for Classification from Aggregate Observations
24. července 2023
Řečníci
Tongliang Liu
Řečník · 0 sledujících
Xiaofeng Zhu
Řečník · 0 sledujících
Heng Tao Shen
Řečník · 0 sledujících
O prezentaci
In conventional supervised classification, true labels are required for individual instances. However, it could be prohibitive to collect the true labels for individual instances, due to privacy concerns or unaffordable annotation costs. This motivates the study on classification from aggregate observations (CFAO), where the supervision is provided to groups of instances, instead of individual instances. CFAO is a generalized learning framework that contains various learning problems, such as multiple-instance learning and learning from label proportions. The goal of this paper is to present a novel universal method of CFAO, which holds an unbiased estimator of the classification risk for arbitrary losses—previous research failed to achieve this goal. Practically, our method works by weighing the importance of each instance and each label in the group, which provides purified supervision for the classifier to learn. Theoretically, our proposed method not only guarantees the risk consistency due to the unbiased risk estimator but also can be compatible with arbitrary losses. Extensive experiments on various problems of CFAO demonstrate the superiority of our proposed method.In conventional supervised classification, true labels are required for individual instances. However, it could be prohibitive to collect the true labels for individual instances, due to privacy concerns or unaffordable annotation costs. This motivates the study on classification from aggregate observations (CFAO), where the supervision is provided to groups of instances, instead of individual instances. CFAO is a generalized learning framework that contains various learning problems, such as mu…
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker
Zhlédnout později
Fast Rates for Maximum Entropy Exploration
Zhlédnout později
Learning Instance-Specific Augmentations by Capturing Local Invariances
Zhlédnout později
Ning Miao, …
End-to-end Differentiable Clustering with Associative Memories
Zhlédnout později
Masked Bayesian Neural Networks : Theoretical Guarantee and its Posterior Inference
Zhlédnout později
Insung Kong, …
RL from Passive Data via Latent Intentions
Zhlédnout později
Dibya Ghosh, …