
PositiveNegative Momentum: Manipulating Stochastic Gradient Noise to Improve Generalization
It is wellknown that stochastic gradient noise (SGN) acts as implicit r...
read it

Approximating InstanceDependent Noise via InstanceConfidence Embedding
Label noise in multiclass classification is a major obstacle to the depl...
read it

Discovering Diverse Solutions in Deep Reinforcement Learning
Reinforcement learning (RL) algorithms are typically limited to learning...
read it

Lowerbounded proper losses for weakly supervised classification
This paper discusses the problem of weakly supervised learning of classi...
read it

LocalDrop: A Hybrid Regularization for Deep Neural Networks
In neural networks, developing regularization algorithms to settle overf...
read it

Incorporating Causal Graphical Prior Knowledge into Predictive Modeling via Simple Data Augmentation
Causal graphs (CGs) are compact representations of the knowledge of the ...
read it

Learning from SimilarityConfidence Data
Weakly supervised learning has drawn considerable attention recently to ...
read it

CIFS: Improving Adversarial Robustness of CNNs via Channelwise Importancebased Feature Selection
We investigate the adversarial robustness of CNNs from the perspective o...
read it

Understanding the Interaction of Adversarial Training with Noisy Labels
Noisy labels (NL) and adversarial examples both undermine trained models...
read it

Learning Noise Transition Matrix from Only Noisy Labels via Total Variation Regularization
Many weakly supervised classification methods employ a noise transition ...
read it

Provably Endtoend LabelNoise Learning without Anchor Points
In labelnoise learning, the transition matrix plays a key role in build...
read it

Learning DiverseStructured Networks for Adversarial Robustness
In adversarial training (AT), the main focus has been the objective and ...
read it

Binary Classification from Multiple Unlabeled Datasets via Surrogate Set Classification
To cope with high annotation costs, training a classifier only from weak...
read it

Sourcefree Domain Adaptation via Distributional Alignment by Matching Batch Normalization Statistics
In this paper, we propose a novel domain adaptation method for the sourc...
read it

A Symmetric Loss Perspective of Reliable Machine Learning
When minimizing the empirical risk in binary classification, it is a com...
read it

Combinatorial Pure Exploration with Fullbandit Feedback and Beyond: Solving Combinatorial Optimization under Uncertainty with Limited Observation
Combinatorial optimization is one of the fundamental research fields tha...
read it

Stable Weight Decay Regularization
Weight decay is a popular regularization technique for training of deep ...
read it

On Focal Loss for ClassPosterior Probability Estimation: A Theoretical Perspective
The focal loss has demonstrated its effectiveness in many realworld app...
read it

Artificial Neural Variability for Deep Learning: On Overfitting, Noise Memorization, and Catastrophic Forgetting
Deep learning is often criticized by two serious issues which rarely exi...
read it

A Survey of Labelnoise Representation Learning: Past, Present and Future
Classical machine learning implicitly assumes that labels of the trainin...
read it

Binary classification with ambiguous training data
In supervised learning, we often face with ambiguous (A) samples that ar...
read it

Classification with Rejection Based on Costsensitive Classification
The goal of classification with rejection is to avoid risky misclassific...
read it

Maximum Mean Discrepancy is Aware of Adversarial Attacks
The maximum mean discrepancy (MMD) test, as a representative twosample ...
read it

Robust Imitation Learning from Noisy Demonstrations
Learning from noisy demonstrations is a practical but highly challenging...
read it

Pointwise Binary Classification with Pairwise Confidence Comparisons
Ordinary (pointwise) binary classification aims to learn a binary classi...
read it

Geometryaware Instancereweighted Adversarial Training
In adversarial machine learning, there was a common belief that robustne...
read it

Provably Consistent PartialLabel Learning
Partiallabel learning (PLL) is a multiclass classification problem, wh...
read it

A Onestep Approach to Covariate Shift Adaptation
A default assumption in many machine learning scenarios is that the trai...
read it

Unbiased Risk Estimators Can Mislead: A Case Study of Learning with Complementary Labels
In weakly supervised learning, unbiased risk estimator(URE) is a powerfu...
read it

Adai: Separating the Effects of Adaptive Learning Rate and Momentum Inertia
Adaptive Momentum Estimation (Adam), which combines Adaptive Learning Ra...
read it

Online Dense Subgraph Discovery via BlurredGraph Feedback
Dense subgraph discovery aims to find a dense component in edgeweighted...
read it

Generalisation Guarantees for Continual Learning with Orthogonal Gradient Descent
In continual learning settings, deep neural networks are prone to catast...
read it

Couplingbased Invertible Neural Networks Are Universal Diffeomorphism Approximators
Invertible neural networks based on coupling flows (CFINNs) have variou...
read it

Analysis and Design of Thompson Sampling for Stochastic Partial Monitoring
We investigate finite stochastic partial monitoring, which is a general ...
read it

LFDProtoNet: Prototypical Network Based on Local Fisher Discriminant Analysis for Fewshot Learning
The prototypical network (ProtoNet) is a fewshot learning framework tha...
read it

Partsdependent Label Noise: Towards Instancedependent Label Noise
Learning with the instancedependent label noise is challenging, because...
read it

Dual T: Reducing Estimation Error for Transition Matrix in Labelnoise Learning
The transition matrix, denoting the transition relationship from clean l...
read it

Similaritybased Classification: Connecting Similarity Learning to Binary Classification
In realworld classification problems, pairwise supervision (i.e., a pai...
read it

Rethinking Importance Weighting for Deep Learning under Distribution Shift
Under distribution shift (DS) where the training data distribution diffe...
read it

Calibrated Surrogate Losses for Adversarially Robust Classification
Adversarially robust classification seeks a classifier that is insensiti...
read it

Learning from Aggregate Observations
We study the problem of learning from aggregate observations where super...
read it

Do Public Datasets Assure Unbiased Comparisons for Registration Evaluation?
With the increasing availability of new image registration approaches, a...
read it

Timevarying Gaussian Process Bandit Optimization with Nonconstant Evaluation Time
The Gaussian process bandit is a problem in which we want to find a maxi...
read it

Attacks Which Do Not Kill Training Make Adversarial Learning Stronger
Adversarial training based on the minimax formulation is necessary for o...
read it

Do We Need Zero Training Loss After Achieving Zero Training Error?
Overparameterized deep networks have the capacity to memorize training d...
read it

Progressive Identification of True Labels for PartialLabel Learning
Partiallabel learning is one of the important weakly supervised learnin...
read it

Towards Mixture Proportion Estimation without Irreducibility
Mixture proportion estimation (MPE) is a fundamental problem of practica...
read it

Fewshot Domain Adaptation by Causal Mechanism Transfer
We study fewshot supervised domain adaptation (DA) for regression probl...
read it

A Diffusion Theory for Deep Learning Dynamics: Stochastic Gradient Descent Escapes From Sharp Minima Exponentially Fast
Stochastic optimization algorithms, such as Stochastic Gradient Descent ...
read it

Learning from Noisy Similar and Dissimilar Data
With the widespread use of machine learning for classification, it becom...
read it
Masashi Sugiyama
is this you? claim profile
Director  RIKEN Center for Advanced Intelligence Project, Professor at University of Tokyo