Liu Ziyin
17 papers · 2020–2025 · 3 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 ↓+7 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (29) π Academic Marathon Β· Sprint π Conference Polyglot Β· Bronze π Interdisciplinary Bridge Β· Bronze π£ Hot Topic Early Bird
π
Keyword Champion
(3)
π
Triple Crown
β‘
Prolific Year Β· Productive
π₯
Unstoppable Β· Bronze
β
The Questioner
π
Century Club Β· Bronze
ποΈ
Keyword Collector
(73)
Venues
Top co-authors
Research topics
Machine Learning
(16)
Optimization & Theory Β· Machine Learning
(14)
Deep Learning
(10)
Artificial Intelligence
(3)
Architectures
(3)
Core AI
(3)
Core Methods
(3)
Optimization & Theory Β· Deep Learning
(3)
Mathematics & Optimization
(2)
Learning Types
(2)
Models
(2)
Techniques
(2)
Computer Vision
(1)
Application Areas
(1)
Bayesian & Probabilistic
(1)
Keywords
stochastic gradient descent
(6)
neural network optimization
(5)
representation learning
(4)
gradient noise
(3)
saddle point
(2)
learning dynamics
(2)
self-supervised learning
(2)
escape rate
(2)
learning rate
(2)
weight decay
(2)
gradient descent
(1)
compositional generalization
(1)
loss landscape
(1)
sparsity
(1)
feature learning
(1)
latent variable model
(1)
contrastive learning
(1)
stationary distribution
(1)
embedding learning
(1)
optimization theory
(1)
Papers
Compositional Generalization via Forced Rendering of Disentangled Latents
ICML 2025
Formation of Representations in Neural Networks
ICLR 2025
Remove Symmetries to Control Model Expressivity and Improve Optimization
ICLR 2025
Three Mechanisms of Feature Learning in a Linear Network
ICLR 2025
Understanding the Emergence of Multimodal Representation Alignment
ICML 2025
Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent
NeurIPS 2024
Symmetry Induces Structure and Constraint of Learning
ICML 2024
On the Stepwise Nature of Self-Supervised Learning
ICML 2023
spred: Solving L1 Penalty with SGD
ICML 2023
What shapes the loss landscape of self supervised learning?
ICLR 2023
Exact Solutions of a Deep Linear Network
NeurIPS 2022
Posterior Collapse of a Linear Latent Variable Model
NeurIPS 2022
Power-Law Escape Rate of SGD
ICML 2022
SGD Can Converge to Local Maxima
ICLR 2022
Strength of Minibatch Noise in SGD
ICLR 2022
Noise and Fluctuation of Finite Learning Rate Stochastic Gradient Descent
ICML 2021
Neural Networks Fail to Learn Periodic Functions and How to Fix It
NeurIPS 2020