Hongzhou Lin
11 papers · 2015–2025 · 6 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 ↓+2 more ↓ Show less ↑
🗺️ Taxonomy Completionist (20) 🐣 Hot Topic Early Bird 🏃 Academic Marathon · Half 🌍 Conference Polyglot · Silver 🌉 Interdisciplinary Bridge · Bronze
❓
The Questioner
💎
Century Club · Bronze
Top co-authors
Research topics
Machine Learning
(11)
Optimization & Theory · Machine Learning
(9)
Mathematics & Optimization
(4)
Optimization
(4)
Artificial Intelligence
(3)
Deep Learning
(2)
Learning Paradigms · Artificial Intelligence
(2)
Application Areas
(1)
Architectures
(1)
Core AI
(1)
Core Methods
(1)
Learning Paradigms · Machine Learning
(1)
Optimization & Theory · Deep Learning
(1)
Keywords
first-order method
(3)
gradient descent
(3)
nesterov acceleration
(2)
non-convex optimization
(2)
convex optimization
(2)
accelerated gradient
(2)
compositional learning
(1)
sample efficiency
(1)
federated learning
(1)
machine unlearning
(1)
task generalization
(1)
decentralized optimization
(1)
knowledge distillation
(1)
worst-case analysis
(1)
representation learning
(1)
strongly convex
(1)
stochastic approximation
(1)
function approximation
(1)
oracle complexity
(1)
variance reduction
(1)
Papers
From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiency
ICLR 2025
Task Generalization with Autoregressive Compositional Structure: Can Learning from D Tasks Generalize to D^T Tasks?
ICML 2025
UNDIAL: Self-Distillation with Adjusted Logits for Robust Unlearning in Large Language Models
NAACL 2025
Beyond Worst-Case Analysis in Stochastic Approximation: Moment Estimation Improves Instance Complexity
ICML 2022
Delayed Gradient Averaging: Tolerate the Communication Latency for Federated Learning
NeurIPS 2021
Complexity of Finding Stationary Points of Nonconvex Nonsmooth Functions
ICML 2020
IDEAL: Inexact DEcentralized Accelerated Augmented Lagrangian Method
NeurIPS 2020
Catalyst Acceleration for First-order Convex Optimization: from Theory to Practice
JMLR 2018
Catalyst for Gradient-based Nonconvex Optimization
AISTATS 2018
ResNet with one-neuron hidden layers is a Universal Approximator
NeurIPS 2018
A Universal Catalyst for First-Order Optimization
NeurIPS 2015