Joint Active Feature Acquisition and Classification with Variable-Size Set Encoding

NeurIPS 2018 Joint Active Feature Acquisition and Classification with Variable-Size Set EncodingHajin Shim, Sung Ju Hwang, Eunho Yang We consider the problem of active feature acquisition, where we sequentially select the subset of features in order to achieve the maximum predict...

DropMax: Adaptive Variational Softmax

NeurIPS 2018 DropMax: Adaptive Variational SoftmaxHaebeom Lee, Juho Lee, Saehoon Kim, Eunho Yang, Sung Ju Hwang We propose DropMax, a stochastic version of softmax classifier which at each iteration drops non-target classes according to dropout probabilities adaptively decided for each ins...

Deep Asymmetric Multi-task Feature Learning

ICML 2018 Deep Asymmetric Multi-task Feature LearningHae Beom Lee, Eunho Yang, Sung Ju Hwang We propose Deep Asymmetric Multitask Feature Learning (Deep-AMTFL) which can learn deep representations shared across multiple tasks while effectively preventing negative transfer that may happen i...

Bucket Renormalization for Approximate Inference

ICML 2018 Bucket Renormalization for Approximate InferenceSungsoo Ahn, Michael Chertkov, Adrian Weller, Jinwoo Shin Probabilistic graphical models are a key tool in machine learning applications. Computing the partition function, i.e., normalizing constant, is a fundamental task of statist...

A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks

NeurIPS 2018 A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial AttacksKimin Lee, Kibok Lee, Honglak Lee, Jinwoo Shin Detecting test samples drawn sufficiently far away from the training distribution statistically or adversarially is a fundamenta...

SplitNet: Learning to Semantically Split Deep Networks for Parameter Reduction and Model Parallelization

ICML 2017 SplitNet: Learning to Semantically Split Deep Networks for Parameter Reduction and Model ParallelizationJuyoung Kim, YooKoon Park, Gunhee Kim, Sungju Hwang We propose a novel deep neural network that is both lightweight and effectively structured for model parallelization. Our ne...

Sparse + Group-Sparse Dirty Models: Statistical Guarantees without Unreasonable Conditions and a Case for Non-Convexity

ICML 2017 Sparse + Group-Sparse Dirty Models: Statistical Guarantees without Unreasonable Conditions and a Case for Non-ConvexityEunho Yang, A. Lozano Imposing sparse + group-sparse superposition structures in high-dimensional parameter estimation is known to provide flexible regularizatio...

Ordinal Graphical Models: A Tale of Two Approaches

ICML 2017 Ordinal Graphical Models: A Tale of Two ApproachesArun Sai Suggala, Eunho Yang, Pradeep Ravikumar Undirected graphical models or Markov random fields (MRFs) are widely used for modeling multivariate probability distributions. Much of the work on MRFs has focused on continuous var...

Learning to Transfer Initializations for Bayesian Hyperparameter Optimization

NeurIPS 2017 Workshop on Bayesian Optimization Learning to Transfer Initializations for Bayesian Hyperparameter OptimizationJungtaek Kim, Saehoon Kim, Seungjin Choi We propose a neural network to learn meta-features over datasets, which is used to select initial points for Bayesian hyperpa...

Graphical Models for Ordinal Data: A Tale of Two Approaches

ICML 2017 Graphical Models for Ordinal Data: A Tale of Two ApproachesArun Sai Suggala, Eunho Yang, P. Ravikumar Undirected graphical models or Markov random fields (MRFs) are widely used for modeling multivariate probability distributions. Much of the work on MRFs has focused on continuous...