conftrace_

Taejong Joo

4 papers · 2020–2024 · 3 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🐣 Hot Topic Early Bird

Venues

ICML (2) ICLR (1) NeurIPS (1)

Top co-authors

Diego Klabjan (2) Byunghoon Kim (1) Min-Gwan Seo (1) Uijung Chung (1) Donggu Kang (1)

Research topics

Machine Learning (4) Deep Learning (2) Application Areas (2) Learning Types · Machine Learning (2) Optimization & Theory (2) Core Methods (1) Learning Types · Deep Learning (1) Techniques (1)

Keywords

distribution shift (2) model calibration (2) bayesian inference (1) uncertainty quantification (1) batch normalization (1) unsupervised domain adaptation (1) label smoothing (1) optimization (1) distribution matching (1) neural network regularization (1) uncertainty estimation (1) pseudo label (1) categorical probability (1) temporal ensemble (1) activation regularization (1) wasserstein distance (1) importance weighted group accuracy estimation (1) model selection (1)

Papers

Improving self-training under distribution shifts via anchored confidence with theoretical guarantees NeurIPS 2024 IW-GAE: Importance weighted group accuracy estimation for improved calibration and model selection in unsupervised domain adaptation ICML 2024 Being Bayesian about Categorical Probability ICML 2020 Regularizing activations in neural networks via distribution matching with the Wasserstein metric ICLR 2020