Min Lin

Principal Research Scientist / Adjunct Assistant Professor

Sea AI Lab

National University of Singapore

News: Sea Industrial PhD Program

Sea AI Lab sponsors Industrial PhD scholarship, apply if you’re interested.

I’m a Principal Research Scientist at the Sea AI Lab, at the same time I’m an Adjunct Assistant Professor at National University of Singapore.

I started my exploration in deep learning around 2013. Started on neural network architecture design, distributed deep learning systems and neural ODE topics during my PhD. Later when I joined Qihoo 360, I applied deep learning on several product lines including search engine, recommendation systems. During my postdoc at Mila, I spent several years working on continual learning (the setting I’m after is actually closer to online learning), and that triggered me to think that we need to go beyond deep learning, i.e. It is not straight-forward how online learning can be achieved with the “data in gradient out” pattern of deep learning.

There are two main themes in my current research at Sea AI Lab,

Exploitation of deep learning to solve problems.
- Systems for ML and ML for Systems.
- Ab Initio Quantum Chemistry methods, including Quantum Monte Carlo and Density Functional Theory.
Exploration of paradigms beyond deep learning.
- I’m optimistic about combining non-parametric methods and generative model to solve online learning problems.

Interests

Ab Initio Quantum Chemistry
ML Systems
Online Learning
Non-parametric Methods
Generative Models
Reverse Engineering

Education

Ph.D. in Electrical and Computer Engineering, 2016

National University of Singapore
B.Sc. in Life Sciences, 2010

Peking University

Experience

Adjunct Assistant Professor

National University of Singapore

Jun 2022 – Present Singapore

Principal Research Scientist

Sea AI Lab (SAIL)

Feb 2021 – Present Singapore

Postdoc

Quebec AI Institute (Mila)

Dec 2017 – Jan 2021 Montreal

Engineer

Qihoo 360 Technology Co. Ltd.

Sep 2015 – Nov 2017 Beijing

Projects

EnvPool

A Highly Parallel Reinforcement Learning Environment Execution Engine

EnvPool

Example Project

An example of using the in-built project page.

Example Project

O(N^2) Universal Antisymmetry in Fermionic Neural Networks

An antisymmetric function approximator for many body fermionic wave function, it is cheaper than Slater Determinant which costs O(N^3).

Publications

Quickly discover relevant content by filtering publications.

Tianyu Pang, Shuicheng Yan, Min Lin (2022). $ O (N^ 2) $ Universal Antisymmetry in Fermionic Neural Networks. arXiv preprint arXiv:2205.13205.

Wenjie Wang, Xinyu Lin, Fuli Feng, Xiangnan He, Min Lin, Tat-Seng Chua (2022). Causal Representation Learning for Out-of-Distribution Recommendation. Proceedings of the ACM Web Conference 2022.

Jiayi Weng, Min Lin, Shengyi Huang, Bo Liu, Denys Makoviichuk, Viktor Makoviychuk, Zichen Liu, Yufan Song, Ting Luo, Yukun Jiang, others (2022). EnvPool: A Highly Parallel Reinforcement Learning Environment Execution Engine. arXiv preprint arXiv:2206.10558.

Tianyu Pang, Min Lin, Xiao Yang, Jun Zhu, Shuicheng Yan (2022). Robustness and Accuracy Could Be Reconcilable by (Proper) Definition. arXiv preprint arXiv:2202.10103.

Xinshuai Dong, Anh Tuan Luu, Min Lin, Shuicheng Yan, Hanwang Zhang (2021). How Should Pre-Trained Language Models Be Fine-Tuned Towards Adversarial Robustness?. Advances in Neural Information Processing Systems.

Xu He, Min Lin (2020). Continual Learning from the Perspective of Compression. arXiv preprint arXiv:2006.15078.

Massimo Caccia, Pau Rodriguez, Oleksiy Ostapenko, Fabrice Normandin, Min Lin, Lucas Caccia, Issam Laradji, Irina Rish, Alexandre Lacoste, David Vazquez, others (2020). Online fast adaptation and knowledge accumulation: a new approach to continual learning. arXiv preprint arXiv:2003.05856.

Min Lin, Jie Fu, Yoshua Bengio (2019). Conditional computation for continual learning. arXiv preprint arXiv:1906.06635.

Rahaf Aljundi, Min Lin, Baptiste Goujaud, Yoshua Bengio (2019). Gradient based sample selection for online continual learning. Advances in neural information processing systems.

Nasim Rahaman, Aristide Baratin, Devansh Arpit, Felix Draxler, Min Lin, Fred Hamprecht, Yoshua Bengio, Aaron Courville (2019). On the spectral bias of neural networks. International Conference on Machine Learning.

Rahaf Aljundi, Eugene Belilovsky, Tinne Tuytelaars, Laurent Charlin, Massimo Caccia, Min Lin, Lucas Page-Caccia (2019). Online continual learning with maximal interfered retrieval. Advances in neural information processing systems.

Nasim Rahaman, Devansh Arpit, Aristide Baratin, Felix Draxler, Min Lin, Fred A Hamprecht, Yoshua Bengio, Aaron C Courville (2018). On the Spectral Bias of Deep Neural Networks..

Min Lin (2017). Softmax gan. arXiv preprint arXiv:1704.06191.

Annan Li, Min Lin, Yi Wu, Ming-Hsuan Yang, Shuicheng Yan (2016). NUS-PRO: A New Visual Tracking Challenge. IEEE transactions on pattern analysis and machine intelligence.

Yunchao Wei, Wei Xia, Min Lin, Junshi Huang, Bingbing Ni, Jian Dong, Yao Zhao, Shuicheng Yan (2015). HCP: A Flexible CNN Framework for Multi-label Image Classification. IEEE Transactions on Pattern Analysis and Machine Intelligence.

Tianqi Chen, Mu Li, Yutian Li, Min Lin, Naiyan Wang, Minjie Wang, Tianjun Xiao, Bing Xu, Chiyuan Zhang, Zheng Zhang (2015). Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems.

Haoqian Zhang, Min Lin, Handuo Shi, Weiyue Ji, Longwen Huang, Xiaomeng Zhang, Shan Shen, Rencheng Gao, Shuke Wu, Chengzhe Tian, others (2014). Programming a Pavlovian-like conditioning circuit in Escherichia coli. Nature communications.

Min Lin, Shuo Li, Xuan Luo, Shuicheng Yan (2014). Purine: A bi-graph based deep learning framework. International Conference on Learning Representations Workshop.

Canyi Lu, Jinhui Tang, Min Lin, Liang Lin, Shuicheng Yan, Zhouchen Lin (2013). Correntropy induced l2 graph for robust subspace clustering. Proceedings of the IEEE International Conference on Computer Vision.

Min Lin, Qiang Chen, Shuicheng Yan (2013). Network In Network. International Conference on Learning Representations.