Kenji Kawaguchi

发表

Discrete-Valued Neural Communication in Structured Architectures Enhances Generalization

Alex Lamb, Chen Sun, Dianbo Liu, 2021 .

Memo No . 90 July 18 , 2019 Theory III : Dynamics and Generalization in Deep Networks 1

T. Poggio, L. Rosasco, B. Miranda, 2019 .

Depth with Nonlinearity Creates No Bad Local Minima in ResNets

Yoshua Bengio, Kenji Kawaguchi, Kenji Kawaguchi, 2018, Neural Networks.

A SIMPLE DEEP EQUILIBRIUM MODEL CONVERGES

Kenji Kawaguchi, Weight Tying, 2021 .

Combined Scaling for Open-Vocabulary Image Classiﬁcation

Quoc V. Le, Adams Wei Yu, Mingxing Tan, 2022 .

Towards Understanding Generalization via Analytical Learning Theory

Yoshua Bengio, L. Kaelbling, Kenji Kawaguchi, 2018, 1802.07426.

Discrete-Valued Neural Communication pdf

Chen Sun, Alex Lamb, Kenji Kawaguchi, 2021, NeurIPS.

When and How Mixup Improves Calibration

James Y. Zou, Linjun Zhang, Zhun Deng, 2021, ICML.

Multi-Task Learning as a Bargaining Game

Ethan Fetaya, Gal Chechik, Kenji Kawaguchi, 2022, ICML.

Two Nicotiana occidentalis accessions enable gene identification for Type II hybrid lethality by the cross to N. sylvestris

Hai He, S. Yokoi, Kenji Kawaguchi, 2021, Scientific Reports.

Sketch-Based Anomaly Detection in Streaming Graphs

Philip S. Yu, Neil Shah, Bryan Hooi, 2021, 2106.04486.

Discrete Key-Value Bottleneck

Anirudh Goyal, B. Scholkopf, Y. Bengio, 2022, ArXiv.

Simplicial Embeddings in Self-Supervised Learning and Downstream Classification

Christos Tsirigotis, A. Vani, Kenji Kawaguchi, 2022, ICLR.

When Do Extended Physics-Informed Neural Networks (XPINNs) Improve Generalization?

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2021, SIAM J. Sci. Comput..

Self-Distillation for Further Pre-training of Transformers

Sung Ju Hwang, Juho Lee, Kenji Kawaguchi, 2022, ArXiv.

Discrete Factorial Representations as an Abstraction for Goal Conditioned Reinforcement Learning

Anirudh Goyal, Riashat Islam, Alex Lamb, 2022, ArXiv.

When Do Extended Physics-Informed Neural Networks (XPINNs) Improve Generalization?

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2021, SIAM J. Sci. Comput..

Neural Active Learning on Heteroskedastic Distributions

Alex Lamb, J. Ash, Cyril Zhang, 2022, ArXiv.

GFlowOut: Dropout with Generative Flow Networks

Bonaventure F. P. Dossou, Chris C. Emezue, Anirudh Goyal, 2022, International Conference on Machine Learning.

Augmented Physics-Informed Neural Networks (APINNs): A gating network-based soft domain decomposition methodology

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2022, ArXiv.

Set-based Meta-Interpolation for Few-Task Meta-Learning

Juho Lee, Kenji Kawaguchi, S. Hwang, 2022, ArXiv.

Understanding Dynamics of Nonlinear Representation Learning and Its Application

Linjun Zhang, Zhun Deng, Kenji Kawaguchi, 2021, Neural Computation.

Every Local Minimum Value Is the Global Minimum Value of Induced Model in Nonconvex Machine Learning

Kenji Kawaguchi, Jiaoyang Huang, Leslie Pack Kaelbling, 2019, Neural Computation.

Discrete Key-Value Bottleneck

Anirudh Goyal, B. Scholkopf, Y. Bengio, 2022, ArXiv.

MemStream: Memory-Based Streaming Anomaly Detection

Bryan Hooi, Kenji Kawaguchi, Siddharth Bhatia, 2021, WWW.

MemStream: Memory-Based Anomaly Detection in Multi-Aspect Streams with Concept Drift pdf

Bryan Hooi, Kenji Kawaguchi, Siddharth Bhatia, 2021, ArXiv.

D4FT: A Deep Learning Approach to Kohn-Sham Density Functional Theory

Min Lin, G. Vignale, A. Neto, 2023, ICLR.

Don’t Just Divide; Polarize and Conquer!

Lingxiao Huang, Kenji Kawaguchi, Siddharth Bhatia, 2021 .

Clustering Aware Classification for Risk Prediction and Subtyping in Clinical Data

Lim Jun Heng, Vaibhav Rajan, Kenji Kawaguchi, 2021, 2102.11872.

CAC: A Clustering Based Framework for Classification pdf

Lingxiao Huang, Kenji Kawaguchi, Siddharth Bhatia, 2021, ArXiv.

Locally adaptive activation functions with slope recovery term for deep and physics-informed neural networks

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2019, ArXiv.

How Does Mixup Help With Robustness and Generalization?

James Y. Zou, Amirata Ghorbani, Linjun Zhang, 2020, ICLR.

GraphMix: Improved Training of GNNs for Semi-Supervised Learning

Juho Kannala, Jian Tang, Kenji Kawaguchi, 2020, AAAI.

Tailoring: encoding inductive biases by optimizing unsupervised objectives at prediction time

L. Kaelbling, Tomas Lozano-Perez, Ferran Alet, 2020, NeurIPS.

Adversarial Training Helps Transfer Learning via Better Representations pdf

James Zou, Zhun Deng, Linjun Zhang, 2021, NeurIPS.

Towards Domain-Agnostic Contrastive Learning

Quoc V. Le, Minh-Thang Luong, Hieu Pham, 2020, ICML.

How Shrinking Gradient Noise Helps the Performance of Neural Networks

Zhun Deng, Kenji Kawaguchi, Jiaoyang Huang, 2021, 2021 IEEE International Conference on Big Data (Big Data).

Ordered SGD: A New Stochastic Optimization Framework for Empirical Risk Minimization

Haihao Lu, Kenji Kawaguchi, 2019, AISTATS.

Meta-learning PINN loss functions pdf

George Em Karniadakis, Kenji Kawaguchi, Apostolos F. Psaros, 2021, J. Comput. Phys..

Optimization of Graph Neural Networks: Implicit Acceleration by Skip Connections and More Depth pdf

Stefanie Jegelka, Keyulu Xu, Kenji Kawaguchi, 2021, ICML.

MixupE: Understanding and Improving Mixup from Directional Derivative Perspective

Hieu Pham, Juho Kannala, Y. Bengio, 2022, arXiv.org.

Automatic Model Selection with Large Language Models for Reasoning

Kenji Kawaguchi, Qizhe Xie, Kenji Kawaguchi, 2023, EMNLP.

Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks

Sung Ju Hwang, Kenji Kawaguchi, Jinheon Baek, 2023, ArXiv.

L AST -L AYER F AIRNESS F INE - TUNING IS S IMPLE AND E FFECTIVE FOR N EURAL N ETWORKS

James Y. Zou, Zhun Deng, Kenji Kawaguchi, 2023 .

Scalable Set Encoding with Universal Mini-Batch Consistency and Unbiased Full Set Gradient Approximation

Juho Lee, Kenji Kawaguchi, S. Hwang, 2022, ICML.

Gradient Descent Finds Global Minima for Generalizable Deep Neural Networks of Practical Sizes

Jiaoyang Huang, Kenji Kawaguchi, Kenji Kawaguchi, 2019, 2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

GFlowOut: Dropout with Generative Flow Networks

Bonaventure F. P. Dossou, Chris C. Emezue, Anirudh Goyal, 2022, ICML.

Multi-View Class Incremental Learning

Cheng Lian, Kenji Kawaguchi, Zhigang Zeng, 2023, 2306.09675.

Set-based Meta-Interpolation for Few-Task Meta-Learning

Juho Lee, Kenji Kawaguchi, S. Hwang, 2022, NeurIPS.

Noether Networks: Meta-Learning Useful Conserved Quantities pdf

Chelsea Finn, Ferran Alet, Joshua Tenenbaum, 2021, NeurIPS.

Deep Kronecker neural networks: A general framework for neural networks with adaptive activation functions

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2021, Neurocomputing.

Discrete Key-Value Bottleneck

B. Schölkopf, Anirudh Goyal, Y. Bengio, 2022, ICML.

EIGNN: Efficient Infinite-Depth Graph Neural Networks

Bryan Hooi, Yiwei Wang, Kenji Kawaguchi, 2022, NeurIPS.

IF2Net: Innately Forgetting-Free Networks for Continual Learning

P. Suganthan, Kenji Kawaguchi, Zhigang Zeng, 2023, ArXiv.

Single-Pass Contrastive Learning Can Work for Both Homophilic and Heterophilic Graph

Jieyu Zhang, Hong Wang, Kenji Kawaguchi, 2022, 2211.10890.

On the Theory of Implicit Deep Learning: Global Convergence with Implicit Layers pdf

Kenji Kawaguchi, Kenji Kawaguchi, 2021, ICLR.

A Recipe for Global Convergence Guarantee in Deep Neural Networks

Qingyun Sun, Kenji Kawaguchi, 2021, AAAI.

MixupE: Understanding and Improving Mixup from Directional Derivative Perspective

Hieu Pham, Juho Kannala, Y. Bengio, 2022, ArXiv.

How Does Information Bottleneck Help Deep Learning?

Xu Ji, Zhun Deng, Kenji Kawaguchi, 2023, ICML.

Last-Layer Fairness Fine-tuning is Simple and Effective for Neural Networks

James Y. Zou, Zhun Deng, Kenji Kawaguchi, 2023, ArXiv.

Adaptive Discrete Communication Bottlenecks with Dynamic Vector Quantization

Alex Lamb, Y. Bengio, Xu Ji, 2022, AAAI.

Locally adaptive activation functions with slope recovery for deep and physics-informed neural networks

Ameya Dilip Jagtap, G. Em Karniadakis, Kenji Kawaguchi, 2020, Proceedings of the Royal Society A.

MGNNI: Multiscale Graph Neural Networks with Implicit Layers

Bryan Hooi, Kenji Kawaguchi, X. Xiao, 2022, NeurIPS.

A Dual-Perspective Approach to Evaluating Feature Attribution Methods

Kenji Kawaguchi, Bernd Bischl, Yanglin Zhang, 2023, ArXiv.

Single-Pass Contrastive Learning Can Work for Both Homophilic and Heterophilic Graph

Jieyu Zhang, Hong Wang, Kenji Kawaguchi, 2022, 2211.10890.

Tackling the Curse of Dimensionality with Physics-Informed Neural Networks

G. Karniadakis, Kenji Kawaguchi, K. Shukla, 2023, ArXiv.

ExpertNet: A Symbiosis of Classification and Clustering pdf

Kenji Kawaguchi, Vaibhav Rajan, Shivin Srivastava, 2022, ArXiv.

Neural Active Learning on Heteroskedastic Distributions

Alex Lamb, J. Ash, Cyril Zhang, 2022, ArXiv.

A Stochastic First-Order Method for Ordered Empirical Risk Minimization pdf

Haihao Lu, Kenji Kawaguchi, Haihao Lu, 2019, ArXiv.

S ELF -S UPERVISED S ET R EPRESENTATION L EARNING FOR U NSUPERVISED M ETA -L EARNING

Kenji Kawaguchi, Seanie Lee, 2023 .

Self-Distillation for Further Pre-training of Transformers

Sung Ju Hwang, Juho Lee, Kenji Kawaguchi, 2022, International Conference on Learning Representations.

Augmented Physics-Informed Neural Networks (APINNs): A gating network-based soft domain decomposition methodology

G. Karniadakis, Ameya Dilip Jagtap, Kenji Kawaguchi, 2022, Eng. Appl. Artif. Intell..

Single-Pass Contrastive Learning Can Work for Both Homophilic and Heterophilic Graph

Jieyu Zhang, Hong Wang, Kenji Kawaguchi, 2022, 2211.10890.

MixupE: Understanding and Improving Mixup from Directional Derivative Perspective

Wai Hoh Tang, Hieu Pham, Juho Kannala, 2022, UAI.

Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks

Sung Ju Hwang, Kenji Kawaguchi, Jinheon Baek, 2023, ArXiv.

GraphMix: Improved Training of GNNs for Semi-Supervised Learning

Alex Lamb, Juho Kannala, Y. Bengio, 2020, AAAI.

Training Free Graph Neural Networks for Graph Matching

Yixin Cao, Xiang Wang, Fuli Feng, 2022, ArXiv.