Kazusato Oko
14 papers · 2022–2025 · 4 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 ↓+5 more ↓ Show less ↑
π£ Hot Topic Early Bird π Conference Polyglot Β· Bronze π Interdisciplinary Bridge Β· Bronze πΊοΈ Taxonomy Completionist (31) π€ Dynamic Duo (14)
π
Triple Crown
β‘
Prolific Year Β· Productive
ποΈ
Keyword Collector
(83)
π
Trend Setter
π
Century Club Β· Bronze
Venues
Top co-authors
Research topics
Machine Learning
(14)
Optimization & Theory Β· Machine Learning
(14)
Deep Learning
(10)
Optimization & Theory Β· Deep Learning
(5)
Architectures
(4)
Core Methods
(4)
Models
(3)
Mathematics & Optimization
(2)
Learning Types Β· Deep Learning
(2)
Optimization
(2)
Artificial Intelligence
(1)
Reinforcement Learning
(1)
Application Areas
(1)
Core AI
(1)
Learning Types Β· Machine Learning
(1)
Keywords
single-index model
(4)
neural network
(4)
sample complexity
(3)
mean-field langevin dynamics
(3)
gradient descent
(2)
in-context learning
(2)
representation learning
(2)
information exponent
(2)
feature learning
(2)
diffusion model
(2)
wasserstein distance
(2)
preference optimization
(1)
mean field theory
(1)
gradient-based optimization
(1)
stochastic gradient descent
(1)
stochastic optimization
(1)
policy gradient
(1)
minimax optimization
(1)
binary classification
(1)
computational complexity
(1)
Papers
Direct Distributional Optimization for Provable Alignment of Diffusion Models
ICLR 2025
Flow matching achieves almost minimax optimal convergence
ICLR 2025
Nonlinear transformers can perform inference-time feature learning
ICML 2025
Improved statistical and computational complexity of the mean-field Langevin dynamics under structured data
ICLR 2024
Learning sum of diverse features: computational hardness and efficient gradient-based training for ridge combinations
COLT 2024
Mean Field Langevin Actor-Critic: Faster Convergence and Global Optimality beyond Lazy Learning
ICML 2024
Neural network learns low-dimensional polynomials with SGD near the information-theoretic limit
NeurIPS 2024
Pretrained Transformer Efficiently Learns Low-Dimensional Target Functions In-Context
NeurIPS 2024
SILVER: Single-loop variance reduction and application to federated learning
ICML 2024
Symmetric Mean-field Langevin Dynamics for Distributional Minimax Problems
ICLR 2024
Diffusion Models are Minimax Optimal Distribution Estimators
ICML 2023
Feature learning via mean-field Langevin dynamics: classifying sparse parities and beyond
NeurIPS 2023
Primal and Dual Analysis of Entropic Fictitious Play for Finite-sum Problems
ICML 2023
Particle Stochastic Dual Coordinate Ascent: Exponential convergent algorithm for mean field neural network optimization
ICLR 2022