Liu Ziyin

发表

On Minibatch Noise: Discrete-Time SGD, Overparametrization, and Bayes pdf

Liu Ziyin, Masahito Ueda, Kangqiao Liu, 2021, ArXiv.

A Simple Approach to the Noisy Label Problem Through the Gambler's Loss

Ruslan Salakhutdinov, Louis-Philippe Morency, Paul Pu Liang, 2019 .

Posterior Collapse of a Linear Latent Variable Model

Liu Ziyin, Zihao Wang, 2022, ArXiv.

Exact Solutions of a Deep Linear Network

Liu Ziyin, Botao Li, Xiangmin Meng, 2022, ArXiv.

Universal Thermodynamic Uncertainty Relation in Non-Equilibrium Dynamics

Liu Ziyin, Masakuni Ueda, 2022, 2202.13311.

Posterior Collapse of a Linear Latent Variable Model

Zihao Wang, Liu Ziyin, 2022, NeurIPS.

Exact Solutions of a Deep Linear Network

Liu Ziyin, Botao Li, Xiangmin Meng, 2022, ArXiv.

Stochastic Neural Networks with Infinite Width are Deterministic

Eric P. Xing, Liu Ziyin, Masakuni Ueda, 2022, ArXiv.

Theoretically Motivated Data Augmentation and Regularization for Portfolio Construction

Liu Ziyin, Kentaro Minami, Kentaro Imajo, 2021, ICAIF.

Price bubbles and policy interventions in the Chinese housing market

Dayong Zhang, Gang-Zhi Fan, Liu Ziyin, 2016, Journal of Housing and the Built Environment.

Sparsity by Redundancy: Solving L1 with a Simple Reparametrization

Zihao Wang, Liu Ziyin, Zihao Wang, 2022, ArXiv.

Neural Networks Fail to Learn Periodic Functions and How to Fix It pdf

Ziyin Liu, Masahito Ueda, Tilman Hartwig, 2020, NeurIPS.

On the Distributional Properties of Adaptive Gradients

Zhang Zhiyi, Liu Ziyin, Liu Ziyin, 2021, UAI.

Logarithmic landscape and power-law escape rate of SGD pdf

Liu Ziyin, Masahito Ueda, Takashi Mori, 2021, ArXiv.

Exact Phase Transitions in Deep Learning

Liu Ziyin, Masakuni Ueda, Li Ziyin, 2022, ArXiv.

What Data Augmentation Do We Need for Deep-Learning-Based Finance? pdf

Liu Ziyin, Kentaro Imajo, Kentaro Minami, 2021, ArXiv.

Multimodal Language Analysis with Recurrent Multistage Fusion

Louis-Philippe Morency, Amir Zadeh, Paul Pu Liang, 2018, EMNLP.

Think Locally, Act Globally: Federated Learning with Local and Global Representations pdf

Ruslan Salakhutdinov, Louis-Philippe Morency, Paul Pu Liang, 2020, ArXiv.

Cross-Modal Generalization: Learning in Low Resource Modalities via Meta-Alignment pdf

Ruslan Salakhutdinov, Louis-Philippe Morency, Paul Pu Liang, 2020, ACM Multimedia.

Deep Gamblers: Learning to Abstain with Portfolio Theory

Ruslan Salakhutdinov, Louis-Philippe Morency, Paul Pu Liang, 2019, NeurIPS.

An Investigation of how Label Smoothing Affects Generalization pdf

Paul Pu Liang, Blair Chen, Ziyin Liu, 2020, ArXiv.

Strength of Minibatch Noise in SGD

Takashi Mori, Liu Ziyin, Masakuni Ueda, 2021, ICLR.

SGD with a Constant Large Learning Rate Can Converge to Local Maxima

James B. Simon, Liu Ziyin, Masakuni Ueda, 2021, 2107.11774.

Noise and Fluctuation of Finite Learning Rate Stochastic Gradient Descent

Liu Ziyin, Masahito Ueda, Kangqiao Liu, 2020, ICML.

SGD May Never Escape Saddle Points

Liu Ziyin, Masakuni Ueda, Botao Li, 2021, ArXiv.

Stochastic Gradient Descent with Large Learning Rate

Masahito Ueda, Liu Ziyin, Kangqiao Liu, 2020, arXiv.org.

LaProp: a Better Way to Combine Momentum with Adaptive Gradient pdf

Ziyin Liu, Masahito Ueda, Zhikang Wang, 2020, ArXiv.

LaProp: Separating Momentum and Adaptivity in Adam

Liu Ziyin, Masahito Ueda, Zhikang T.Wang, 2020, 2002.04839.

What shapes the loss landscape of self-supervised learning?

Hidenori Tanaka, Liu Ziyin, Masakuni Ueda, 2022, ICLR.

Exact Solutions of a Deep Linear Network

Liu Ziyin, Botao Li, Xiangmin Meng, 2022, NeurIPS.

Power-Law Escape Rate of SGD

Takashi Mori, Liu Ziyin, Kangqiao Liu, 2021, ICML.

SGD Can Converge to Local Maxima

James B. Simon, Masahito Ueda, Liu Ziyin, 2022, ICLR.

Learning Not to Learn in the Presence of Noisy Labels pdf

Ruslan Salakhutdinov, Louis-Philippe Morency, Paul Pu Liang, 2020, ArXiv.

spred: Solving L1 Penalty with SGD

Liu Ziyin, Zihao Wang, 2022, ICML.

Exact solutions of a deep linear network

Liu Ziyin, Botao Li, Xiangmin Meng, 2022, NeurIPS.

Balancing sensory inputs: somatosensory reweighting from proprioception to tactile sensation in maintaining postural stability among older adults with sensory deficits

Qipeng Song, Wei Sun, Liu Ziyin, 2023, Frontiers in Public Health.

On the stepwise nature of self-supervised learning

James B. Simon, Joshua Albrecht, Liu Ziyin, 2023, ICML.

spred: Solving L1 Penalty with SGD

Liu Ziyin, Zihao Wang, 2022, International Conference on Machine Learning.

The Probabilistic Stability of Stochastic Gradient Descent

Tomer Galanti, Liu Ziyin, Masakuni Ueda, 2023, ArXiv.

Law of Balance and Stationary Distribution of Stochastic Gradient Descent

Liu Ziyin, Masakuni Ueda, Hongchao Li, 2023, ArXiv.