Hanchen Wang

Hey! I am a third-year PhD in Machine Learning at Cambridge, working with Prof. Joan Lasenby from Trinity. During PhD, I also spend some time at Isaac Newton Institute, Cambridge Math, Google Research, Amazon Alexa, BioMap, ZhenFund, Entos, and Caltech (with Prof. Anima) [where I live].

I earned a Physics BS from Nanjing University (2018), visited Stanford & Cal, worked in Finance (S&T, Quant). I've conducted research projects with Prof. Xinran Wang and Prof. Ali Javey on electronic devices and solar cells.

I was admitted to University at age 15, and later served as the Valedictorian.

CV |  Email  |  Twitter |  Google Scholar

profile photo
Interests in Research [in Life] >

I work on 3D, Graph and Imaging data, now emphasizing Genomics (scRNA seq, Perturb-Seq, Multi-Modal).
I also spent some time on Machine Learning System, Quantum Computation, and Geometry.
I co-lead large initiatives also advance terse yet forceful techniques.

Updates [More] >

[22.09] heading to Genentech & Stanford in Bay Area for a Superday!
[22.07] physically co-organizing the 2nd AI4Science workshop at ICML '22
[22.05] moving to La Jolla and Arcadia, starting my one-year journey at Entos and Caltech

Selected Works [Full List] >
* denotes the joint first authorship

Enabling Scientific Discovery with Artificial Intelligence
Hanchen Wang*, et al., [Authoring team] >
in submission, for Nature
Main Figure
Predicting Biomarkers from Histopathological Images via Saliency Lesion Searching
Jiefeng Gan*, Hanchen Wang*, et al., [Authoring team] >
in review, for Nature Communications
Main Figure
Open Problems in Federated Digital Health
Hanchen Wang, James Zou
revising, for Nature Machine Intelligence
Evaluating Self-supervised Learning for Molecular Graph Embeddings
Hanchen Wang*, Jean Kaddour*, Shengchao Liu, Jian Tang, Matt Kusner, Joan Lasenby, Qi Liu
in review, also presented at Pre-Training & AI4Science at ICML 2022
interesting reviews on OpenReview
Pre-training Molecular Graph Representation with 3D Geometry
Shengchao Liu, Hanchen Wang, Weiyang Liu, Joan Lasenby, Hongyu Guo, Jian Tang
ICLR 2022, also presented at SSL NeurIPS 2021, GTRL ICLR 2022 (Spotlight)
project page  |  code  |  bibtex
Matching Point Sets with Quantum Circuits Learning
H. Wang*, M. Noormandipour* (in contribution order)
ICASSP 2022, Invited by Editors, with Travel Award
project page  |  code  |  bibtex
Iterative Teaching by Label Synthesis
Weiyang Liu*, Zhen Liu*, Hanchen Wang*, Liam Paul, Bernhard Schölkopf, Adrian Weller
NeurIPS 2021, Spotlight
OpenReview  |  bibtex
Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence
Xiang Bai* Hanchen Wang*, Liya Ma*, Yongchao Xu*, Jiefeng Gan*, ..., Carola Schönlieb, Tian Xia
Nature Machine Intelligence 2021
project page  |  code  |  bibtex-paper  |  bibtex-code
Media coverage: HUST, Cambridge, Horizon Magazine, Tech Xplore, MIT Technology Review, Synced, [upcoming...]
Unsupervised Point Cloud Pre-training via Occlusion Completion
Hanchen Wang, Qi Liu, Xiangyu Yue, Joan Lasenby, Matthew J. Kusner
ICCV 2021, also presented at WeaSul ICLR 2021
project page  |  code  |  bibtex
Negative Capacitance 2D MoS2 Transistors with Sub-60mV/dec Subthreshold Swing over 6 Orders,
250 μA/μm Current Density, and Nearly-Hysteresis-Free

Zhihao Yu*, Hanchen Wang*, ..., Xinran Wang
IEDM 2017, Oral

Open Source Contribution [GitHub] >

deeplearning.ai: give internal feedback on the newly publicized courses on Natural Language Processing and TensorFlow2.

PyTorch3D: A library for 3D deep learning developed by talents from Facebook AI Research.

metric-learn: Supervised and weakly-supervised Mahalanobis metric learning algorithms.

openmlsys-zh: MLSys Textbook.

Teaching Assistant
- Inference, Lent 2020 and 2021
- Lab IB: Spectrum Analysis, Michaelmas 2020
- Statistical Signal Processing, Michaelmas 2019 and 2020

- Workshop Proposals for NeurIPS, ICML
- ICLR, ICML, NeurIPS, CVPR, KDD, AAAI, and so on
- Nature Machine Intelligence, Nature Electronic, TPAMI, and so on

- 1st/2nd/3rd Workshops on "AI for Science", NeurIPS '21/ICML '22/NeurIPS '22, Twitter, LinkedIn
        sponsored by Microsoft Research Cambridge/Amsterdam, BioMap and DeepMind.
- Workshop on "ML for Genomes", in preparation

- Puria Radmard, InfoEng, Trinity '22, "Generative Models on Satellite images"
- a Group of Berkeley EECS MEng '22, "Continual Segmentation Adaptation"

Talks [More] >
Research, Startup
- [18-21] quite a bite on some QishiCPC (Premium Member) seminars
- [19. xx] call/pitch/pres on ''Cantab Care'', a startup I co-founded

- [22.07] Overview on DepMap, Potluck Seminar Group
- [22.07] Evaluating Self-Supervised Learned Molecular Graphs, (ex) Miller Group, Caltech
- [22.06] Pre-training Molecular Graphs with 3D Geometry, Lennard-Jones Centre, Cam
- [22.02] Rethinking Self-Supervised Learning on Structured Data, ML/NLP Oxford
- [21.10] Graph Denoising via Edge Editing, Amazon Machine Learning Conference

- Role models: Neil Shen, Kai-Fu Lee
- Love Reading, Noting, Travelling, Cooking & Exercising
- Achieved the 3rd level of National Athlete in 400 metres
- Know a whit about Classical Chinese, Japanese and French
- Used to be Vice President of Cambridge Algorithmic Trading Society
- Won the 2nd prizes in both China High School Biology and Physics Olympiad
- Used to play instruments incl. Erhu (10 years) and Ocarina, achieved the highest level at 11
- Trade on Crypto (since '19) and US/CN Equity (since '18), sometimes in automated high-freq manners

template is based on brilliant Barron's
last modified in 22.06