FedMix: Approximation of Mixup under Mean Augmented Federated Learning

ICLR 2021 FedMix: Approximation of Mixup under Mean Augmented Federated Learning Tehrim Yoon, Sumin Shin, Sung Ju Hwang and Eunho Yang  Federated learning (FL) allows edge devices to collectively learn a model without directly sharing data within each device, thus preserving priv...

Federated Semi-Supervised Learning with Inter-Client Consistency & Disjoint Learning

ICLR 2021 (formerly ICML 2020 Workshop - Best Student Paper Award) Federated Semi-Supervised Learning with Inter-Client Consistency & Disjoint Learning Wonyong Jeong, Jaehong Yoon, Eunho Yang and Sung Ju Hwang  While existing federated learning approaches mostl...

Diversity Matters When Learning From Ensembles

ICLR 2021 Diversity Matters When Learning From Ensembles  Giung Nam, Jongmin Yoon, Yoonho Lee, Juho Lee   Deep ensembles excel in large-scale image classification tasks both in terms of prediction accuracy and calibration. Despite being simple to train, the computa...

Contrastive Learning with Adversarial Perturbations for Conditional Text Generation

ICLR 2021 Contrastive Learning with Adversarial Perturbations for Conditional Text Generation Seanie Lee, Dong Bok Lee and Sung Ju Hwang  Recently, sequence-to-sequence (seq2seq) models with the Transformer architecture have achieved remarkable performance on various conditional ...

Accurate Learning of Graph Representations with Graph Multiset Pooling

ICLR 2021 Accurate Learning of Graph Representations with Graph Multiset Pooling Jinheon Baek, Minki Kang and Sung Ju Hwang  Message-passing graph neural networks have been widely used on modeling graph data, achieving impressive results on a number of graph classification and li...

Why Not to Use Zero Imputation? Correcting Sparsity Bias in Training Neural Networks

ICLR 2020 Why Not to Use Zero Imputation? Correcting Sparsity Bias in Training Neural Networks Joonyoung Yi, Juhyuk Lee, Kwang Joon Kim, Sung Ju Hwang, Eunho Yang  Handling missing data is one of the most fundamental problems in machine learning. Among many approaches, ...

Scalable and Order-robust Continual Learning with Additive Parameter Decomposition

ICLR 2020 Scalable and Order-robust Continual Learning with Additive Parameter Decomposition Jaehong Yoon, Saehoon Kim, Eunho Yang, Sung Ju Hwang  While recent continual learning methods largely alleviate the catastrophic problem on toy-sized datasets, there are issues that ...

Meta Dropout: Learning to Perturb Latent Features for Generalization

ICLR 2020 Meta Dropout: Learning to Perturb Latent Features for Generalization Hae Beom Lee, Taewook Nam, Eunho Yang, Sung Ju Hwang  A machine learning model that generalizes well should obtain low errors on unseen test examples. Thus, if we know how to optimally perturb tra...

Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Tasks

ICLR 2020 Learning to Balance: Bayesian Meta-Learning forImbalanced and Out-of-distribution Tasks Hae Beom Lee, Hayeon Lee, Donghyun Na, Saehoon Kim, Minseop Park, Eunho Yang, Sung Ju Hwang  While tasks could come with varying the number of instances and classes in realistic sett...

Learning to Propagate Labels: Transductive Propagation Network for Few-shot Learning

ICLR 2019 Learning to Propagate Labels: Transductive Propagation Network for Few-shot LearningYanbin Liu, Juho Lee, Minseop Park, Saehoon Kim, Eunho Yang, Sung Ju Hwang, Yi Yang The goal of few-shot learning is to learn a classifier that generalizes well even when trained with a ...