Liu Ziyin - RankMe

60 papers A* 18Journal 41Unranked 1

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Does SGD Seek Flatness or Sharpness? An Exactly Solvable Model. CoRR Yizhou Xu, Pierfrancesco Beneventano, Isaac L. Chuang, Liu Ziyin
2025	J	`jnl`	A universal compression theory: Lottery ticket hypothesis and superpolynomial scaling laws. CoRR Hong-Yi Wang, Di Luo, Tomaso A. Poggio, Isaac L. Chuang, Liu Ziyin
2025	J	`jnl`	An Equivariance Toolbox for Learning Dynamics. CoRR Yongyi Yang, Liu Ziyin
2025	J	`jnl`	Compositional Generalization Requires More Than Disentangled Representations. CoRR Qiyao Liang, Daoyuan Qian, Liu Ziyin, Ila Fiete
2025	A*	`conf`	Compositional Generalization via Forced Rendering of Disentangled Latents. ICML Qiyao Liang, Daoyuan Qian, Liu Ziyin, Ila R. Fiete
2025	J	`jnl`	Emergence of Hebbian Dynamics in Regularized Non-Local Learners. CoRR David Koplow, Tomaso A. Poggio, Liu Ziyin
2025	A*	`conf`	Formation of Representations in Neural Networks. ICLR Liu Ziyin, Isaac L. Chuang, Tomer Galanti, Tomaso A. Poggio
2025	J	`jnl`	Heterosynaptic Circuits Are Universal Gradient Machines. CoRR Liu Ziyin, Isaac L. Chuang, Tomaso A. Poggio
2025	J	`jnl`	Neural Thermodynamics I: Entropic Forces in Deep and Universal Representation Learning. CoRR Liu Ziyin, Yizhou Xu, Isaac L. Chuang
2025	J	`jnl`	On Biologically Plausible Learning in Continuous Time. CoRR Marc Gong Bacvanski, Liu Ziyin, Tomaso A. Poggio
2025	J	`jnl`	Parameter Symmetry Breaking and Restoration Determines the Hierarchical Learning in AI Systems. CoRR Liu Ziyin, Yizhou Xu, Tomaso A. Poggio, Isaac L. Chuang
2025	A*	`conf`	Remove Symmetries to Control Model Expressivity and Improve Optimization. ICLR Liu Ziyin, Yizhou Xu, Isaac L. Chuang
2025	J	`jnl`	Topological Invariance and Breakdown in Learning. CoRR Yongyi Yang, Tomaso A. Poggio, Isaac L. Chuang, Liu Ziyin
2025	A*	`conf`	Understanding the Emergence of Multimodal Representation Alignment. ICML Megan Tjandrasuwita, Chanakya Ekbote, Liu Ziyin, Paul Pu Liang
2025	J	`jnl`	Understanding the Emergence of Multimodal Representation Alignment. CoRR Megan Tjandrasuwita, Chanakya Ekbote, Liu Ziyin, Paul Pu Liang
2025	J	`jnl`	When Reasoning Meets Its Laws. CoRR Junyu Zhang, Yifan Sun, Tianang Leng, Jingyan Shen, Liu Ziyin, Paul Pu Liang, Huan Zhang
2024	J	`jnl`	Formation of Representations in Neural Networks. CoRR Liu Ziyin, Isaac L. Chuang, Tomer Galanti, Tomaso A. Poggio
2024	A*	`conf`	Parameter Symmetry and Noise Equilibrium of Stochastic Gradient Descent. NeurIPS Liu Ziyin, Mingze Wang, Hongchao Li, Lei Wu
2024	J	`jnl`	Remove Symmetries to Control Model Expressivity. CoRR Liu Ziyin, Yizhou Xu, Isaac L. Chuang
2024	J	`jnl`	Self-Assembly of a Biologically Plausible Learning Circuit. CoRR Qianli Liao, Liu Ziyin, Yulu Gan, Brian Cheung, Mark Harnett, Tomaso A. Poggio
2024	A*	`conf`	Symmetry Induces Structure and Constraint of Learning. ICML Liu Ziyin
2024	J	`jnl`	The Implicit Bias of Gradient Noise: A Symmetry Perspective. CoRR Liu Ziyin, Mingze Wang, Lei Wu
2024	J	`jnl`	When Does Feature Learning Happen? Perspective from an Analytically Solvable Model. CoRR Yizhou Xu, Liu Ziyin
2023	J	`jnl`	Law of Balance and Stationary Distribution of Stochastic Gradient Descent. CoRR Liu Ziyin, Hongchao Li, Masahito Ueda
2023	A*	`conf`	On the Stepwise Nature of Self-Supervised Learning. ICML James B. Simon, Maksis Knutins, Liu Ziyin, Daniel Geisz, Abraham J. Fetterman, Joshua Albrecht
2023	J	`jnl`	On the stepwise nature of self-supervised learning. CoRR James B. Simon, Maksis Knutins, Liu Ziyin, Daniel Geisz, Abraham J. Fetterman, Joshua Albrecht
2023	J	`jnl`	Symmetry Leads to Structured Constraint of Learning. CoRR Liu Ziyin
2023	J	`jnl`	The Probabilistic Stability of Stochastic Gradient Descent. CoRR Liu Ziyin, Botao Li, Tomer Galanti, Masahito Ueda
2023	A*	`conf`	What shapes the loss landscape of self supervised learning? ICLR Liu Ziyin, Ekdeep Singh Lubana, Masahito Ueda, Hidenori Tanaka
2023	A*	`conf`	spred: Solving L1 Penalty with SGD. ICML Liu Ziyin, Zihao Wang
2022	J	`jnl`	Exact Phase Transitions in Deep Learning. CoRR Liu Ziyin, Masahito Ueda
2022	A*	`conf`	Exact Solutions of a Deep Linear Network. NeurIPS Liu Ziyin, Botao Li, Xiangming Meng
2022	J	`jnl`	Exact Solutions of a Deep Linear Network. CoRR Liu Ziyin, Botao Li, Xiangming Meng
2022	A*	`conf`	Posterior Collapse of a Linear Latent Variable Model. NeurIPS Zihao Wang, Liu Ziyin
2022	J	`jnl`	Posterior Collapse of a Linear Latent Variable Model. CoRR Zihao Wang, Liu Ziyin
2022	A*	`conf`	Power-Law Escape Rate of SGD. ICML Takashi Mori, Liu Ziyin, Kangqiao Liu, Masahito Ueda
2022	A*	`conf`	SGD Can Converge to Local Maxima. ICLR Liu Ziyin, Botao Li, James B. Simon, Masahito Ueda
2022	J	`jnl`	Sparsity by Redundancy: Solving L CoRR Liu Ziyin, Zihao Wang
2022	J	`jnl`	Stochastic Neural Networks with Infinite Width are Deterministic. CoRR Liu Ziyin, Hanlin Zhang, Xiangming Meng, Yuting Lu, Eric P. Xing, Masahito Ueda
2022	A*	`conf`	Strength of Minibatch Noise in SGD. ICLR Liu Ziyin, Kangqiao Liu, Takashi Mori, Masahito Ueda
2022	—	`conf`	Theoretically Motivated Data Augmentation and Regularization for Portfolio Construction. ICAIF Liu Ziyin, Kentaro Minami, Kentaro Imajo
2022	J	`jnl`	What shapes the loss landscape of self-supervised learning? CoRR Liu Ziyin, Ekdeep Singh Lubana, Masahito Ueda, Hidenori Tanaka
2021	A*	`conf`	Cross-Modal Generalization: Learning in Low Resource Modalities via Meta-Alignment. ACM Multimedia Paul Pu Liang, Peter Wu, Liu Ziyin, Louis-Philippe Morency, Ruslan Salakhutdinov
2021	J	`jnl`	Logarithmic landscape and power-law escape rate of SGD. CoRR Takashi Mori, Liu Ziyin, Kangqiao Liu, Masahito Ueda
2021	A*	`conf`	Noise and Fluctuation of Finite Learning Rate Stochastic Gradient Descent. ICML Kangqiao Liu, Liu Ziyin, Masahito Ueda
2021	J	`jnl`	On Minibatch Noise: Discrete-Time SGD, Overparametrization, and Bayes. CoRR Liu Ziyin, Kangqiao Liu, Takashi Mori, Masahito Ueda
2021	J	`jnl`	On the Distributional Properties of Adaptive Gradients. CoRR Zhiyi Zhang, Liu Ziyin
2021	J	`jnl`	SGD May Never Escape Saddle Points. CoRR Liu Ziyin, Botao Li, Masahito Ueda
2021	J	`jnl`	What Data Augmentation Do We Need for Deep-Learning-Based Finance? CoRR Liu Ziyin, Kentaro Minami, Kentaro Imajo
2020	J	`jnl`	An Investigation of how Label Smoothing Affects Generalization. CoRR Blair Chen, Liu Ziyin, Zihao Wang, Paul Pu Liang
2020	J	`jnl`	Cross-Modal Generalization: Learning in Low Resource Modalities via Meta-Alignment. CoRR Paul Pu Liang, Peter Wu, Liu Ziyin, Louis-Philippe Morency, Ruslan Salakhutdinov
2020	J	`jnl`	LaProp: a Better Way to Combine Momentum with Adaptive Gradient. CoRR Liu Ziyin, Zhikang Wang, Masahito Ueda
2020	J	`jnl`	Learning Not to Learn in the Presence of Noisy Labels. CoRR Liu Ziyin, Blair Chen, Ru Wang, Paul Pu Liang, Ruslan Salakhutdinov, Louis-Philippe Morency, Masahito Ueda
2020	A*	`conf`	Neural Networks Fail to Learn Periodic Functions and How to Fix It. NeurIPS Liu Ziyin, Tilman Hartwig, Masahito Ueda
2020	J	`jnl`	Neural Networks Fail to Learn Periodic Functions and How to Fix It. CoRR Liu Ziyin, Tilman Hartwig, Masahito Ueda
2020	J	`jnl`	Stochastic Gradient Descent with Large Learning Rate. CoRR Kangqiao Liu, Liu Ziyin, Masahito Ueda
2020	J	`jnl`	Think Locally, Act Globally: Federated Learning with Local and Global Representations. CoRR Paul Pu Liang, Terrance Liu, Liu Ziyin, Ruslan Salakhutdinov, Louis-Philippe Morency
2020	J	`jnl`	Volumization as a Natural Generalization of Weight Decay. CoRR Liu Ziyin, Zihao Wang, Makoto Yamada, Masahito Ueda
2019	A*	`conf`	Deep Gamblers: Learning to Abstain with Portfolio Theory. NeurIPS Liu Ziyin, Zhikang Wang, Paul Pu Liang, Ruslan Salakhutdinov, Louis-Philippe Morency, Masahito Ueda
2019	J	`jnl`	Deep Gamblers: Learning to Abstain with Portfolio Theory. CoRR Liu Ziyin, Zhikang Wang, Paul Pu Liang, Ruslan Salakhutdinov, Louis-Philippe Morency, Masahito Ueda