Zhouchen Lin
教授
Researches machine learning, numerical optimization, pattern recognition and computer vision.
Zhouchen Lin's lab at Peking works on learning theory & optimization and recognition, detection & segmentation. 122 papers since 2023, 47 at top venues, mostly ICML and ICLR. 17 PhD students and 5 master's students. 41 alumni. Applications close .
At a glance
- papers, 2023–2026
- 32/ year
- at T1 venues
- 39%
- PhD students
- 17
- MS students
- 5
- alumni listed
- 41
Papers by yearfrom the lab homepage, 2023–2026
- 37 papers2026 · T1 20 · collecting
- 37 papers2025 · T1 18
- 25 papers2024 · T1 4
- 23 papers2023 · T1 5
T1 venuesOtherVenues: ICML, ICLR, NeurIPS, Neural Networks, COLT
Research focuspaper titles that use each topic's words
- Learning theory & optimization23 papers
- Recognition, detection & segmentation3 papers
- numerical0 papers
- intelligence science0 papers
- pattern recognition0 papers
Paper list
From the lab homepage.
202637 papers
- ACM Conference on Recommender Systems (RecSys ), Industry Track (Oral).A Self-Triggered Agentic Push Recommendation System (opens in a new tab)
- IEEE Trans. Pattern Analysis and Machine Intelligence, 48(2): 1521-1541 .A Survey on Video Temporal Grounding with Multimodal Large Language Model (opens in a new tab)
- IEEE Trans. Geoscience and Remote SensingAdaMixCon: Difficulty-Aware Feature Enhancement for Few-Shot Hyperspectral Image Classification (opens in a new tab)
- Anda Tang and Zhouchen Lin*, 大模型参数高效微调方法综述：技术、趋势与挑战，自动化学报，pp. 1520-1556，Aug. .Anda Tang and Zhouchen Lin* (opens in a new tab)
- NeurIPS (T1)Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers (opens in a new tab)
- Machine LearningBeyond Likelihood Ratio Bias: Nested Multi-Time-Scale Stochastic Approximation for Likelihood-Free Parameter Estimation (opens in a new tab)
- ICML (T1)Beyond Rational Illusion: Behaviorally Realistic Strategic Classification (opens in a new tab)
- Nature Machine Intelligence (T1)Beyond Representational Alignment: Brain-guided Language Models for Robust Reasoning (opens in a new tab)
- ICML (T1)Convergence Rate Analysis of the AdamW-Style Shampoo: Unifying One-sided and Two-Sided Preconditioning (opens in a new tab)
- SIGIR (T1)Debiased Recommendation Beyond the Positive Propensity Assumption (opens in a new tab)
- ICLR (T1)DistDF: Time-series Forecasting Needs Joint-distribution Wasserstein Alignment (opens in a new tab)
- ICLR (T1)DNT: a Deeply Normalized Transformer That Can be Trained by Momentum SGD (opens in a new tab)
202537 papers
- ICML (T1)Active Treatment Effect Estimation via Limited Samples (opens in a new tab)
- NeurIPS (T1)AdaMSS: Adaptive Multi-Subspace Approach for Parameter-Efficient Fine-Tuning (opens in a new tab)
- ICLR (T1)Affine Steerable Equivariant Layer for Canonicalization of Neural Networks (opens in a new tab)
- SIGIR : 895-905.AV-NAS: Audio-Visual Multi-Level Semantic Neural Architecture Search for Video Hashing (opens in a new tab)
- IEEE Trans. Pattern Analysis and Machine Intelligence, 47(11): 9688-9701 .ClusMatch: Improving Deep Clustering by Unified Positive and Negative Pseudo-label Learning (opens in a new tab)
- (in Chinese), Chinese J. ComputersConvergence Rate Analysis of LION (opens in a new tab)
- IJCAI : 10400-10408, survey track.Empowering LLMs with Logical Reasoning: A Comprehensive Survey (opens in a new tab)
- ICML (T1)Explicit Discovery of Nonlinear Symmetries from Dynamic Data (opens in a new tab)
- ICML AI for Math Workshop.Governing Equation Discovery from Data Based on Differential Invariants (opens in a new tab)
- J. Machine Learning Research, 26 1-53.High-Rank Irreducible Cartesian Tensor Decomposition and Bases of Equivariant Spaces (opens in a new tab)
- Big Data and Information AnalyticsHuman Structure Modeling for Video-based Person Re-identification without Body-part Labels (opens in a new tab)
- NeurIPS (T1)Improving Model Representation and Reducing KV Cache via Skip Connections with First Value Heads (opens in a new tab)
202425 papers
- J. Machine Learning Research, 25(274):1-52, .Accelerated Gradient Tracking over Time-varying Graphs for Decentralized Optimization (opens in a new tab)
- IEEE Trans. Pattern Analysis and Machine Intelligence, 46(12): 9508-9520 . [full version]Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models (opens in a new tab)
- CVPR . [supplementary material]Affine Equivariant Networks Based on Differential Invariants (opens in a new tab)
- National Science Review, 11: nwad292, .Bilevel Optimization for Automated Machine Learning: A New Perspective on Framework and Algorithm (opens in a new tab)
- Biomedical Signal Processing and ControlCross-Patient Automatic Epileptic Seizure Detection Using Patient-Adversarial Neural Networks with Spatio-Temporal EEG Augmentation (opens in a new tab)
- IEEE Trans. Pattern Analysis and Machine Intelligence, 46(9): 6231-6246 .Designing Universally-Approximating Deep Neural Networks: A First-Order Optimization Approach (opens in a new tab)
- Frontiers in Neuroscience, section Neuromorphic Engineering, 17:1303564 .Efficient and Generalizable Cross-patient Epileptic Seizure Detection through a Spiking Neural Network (opens in a new tab)
- Neural NetworksEfficient Learning of Scale-Adaptive Nearly Affine Invariant Networks (opens in a new tab)
- SCIENCE CHINA Information SciencesEmbodied Computational Imaging: A New Paradigm for Observing and Analyzing Spatiotemporally Ultrasensitive Phenomena at Multiple Scales (opens in a new tab)
- ICLR (T1)Hebbian Learning based Orthogonal Projection for Continual Learning of Spiking Neural Networks (opens in a new tab)
- AAAI (T1)LION: Implicit Vision Prompt Tuning (opens in a new tab)
- ECAI .MixCon: A Hybrid Architecture for Efficient and Adaptive Sequence Modeling (opens in a new tab)
202323 papers
- ICLR (T1)A Message Passing Perspective on Learning Dynamics of Contrastive Learning (opens in a new tab)
- NeurIPS (oral). [supplementary material]A Single-Loop Accelerated Extra-Gradient Difference Algorithm with Improved Complexity Bounds for Constrained Minimax Optimization (opens in a new tab)
- NeurIPS . [supplementary material]Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective (opens in a new tab)
- IEEE Trans. Image Processing, 32: 1868-1881 .Equilibrium Image Denoising with Implicit Differentiation (opens in a new tab)
- NeurIPS . [supplementary material]GEQ: Gaussian Kernel Inspired Equilibrium Models (opens in a new tab)
- AISTATS . [supplementary material]Global Convergence of Over-parameterized Deep Equilibrium Models (opens in a new tab)
- IJCNNGradient Descent Optimizes Normalization-Free ResNets (opens in a new tab)
- s, International J. Computer Vision, 131(1): 134-159 .Invertible Rescaling Network and Its Extension (opens in a new tab)
- IJCAI (T1)KDLGT: A Linear Graph Transformer Framework via Kernel Decomposition Approach (opens in a new tab)
- ICLR (T1)Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning (opens in a new tab)
- ICLR . [supplementary material]Neural ePDOs: Spatially Adaptive Equivariant Partial Differential Operator Based Networks (opens in a new tab)
- SCIENCE CHINA Information SciencesNewton Design: Designing CNNs with the Family of Newton’s Methods (opens in a new tab)
Where alumni went41 listed
Apply to PekingFall 2027
Computer Science PhD (international applicants)
- Round
- Main
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- CNZhejiang
Fei Wu
- Knowledge, QA & retrieval
- USGeorgia Tech
Humphrey Shi
- Image & video generation
- CNZhejiang
Jianke Zhu
- Recommendation, search & web
- CNPeking
Yizhou Wang
- Reasoning & agents
- USUIUC
Alexander Schwing
- Vision-language & multimodal
- USNYU
Saining Xie
- Vision-language & multimodal
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- This site is shared with other labs, so counts may overlap.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
This lab shares a website with other labs. Member and alumni counts may include them.
Tiers are venue classifications, not lab rankings.
Pekingcloses