Wenlin Chen

PhD Student in Machine Learning

University of Cambridge

Max Planck Institute for Intelligent Systems

About Me

I am a PhD student in Machine Learning at University of Cambridge (Machine Learning Group, Computational and Biological Learning Lab) and Max Planck Institute for Intelligent Systems (Empirical Inference Department), under the Cambridge-Tübingen PhD Fellowship in Machine Learning. My supervisors are Professor José Miguel Hernández-Lobato and Professor Bernhard Schölkopf. My advisor is Dr Hong Ge.

I am keen on basic research in machine learning and its scientific applications (e.g., drug discovery). My research interest lies at the intersection of probabilistic methods, deep learning, and causal inference. I aim to develop data-efficient and causally-aware machine learning methods that enable active data collection, robust inference and prediction, efficient data compression, and realistic generation of novel synthetic data.

Interests

Probabilistic Modelling
Deep Learning
Representation Learning
Multi-task/Meta Learning
Generative Modelling
Optimization
Causal Modelling

Education

PhD in Machine Learning

University of Cambridge and Max Planck Institute for Intelligent Systems
MPhil in Machine Learning and Machine Intelligence, Distinction, 2021

University of Cambridge
BSc in Mathematics, First Class Honours, 2020

University of Manchester

Publications

See also my Google Scholar profile or search on my publication page.

Wenlin Chen, Hong Ge (2023). Neural Characteristic Activation Value Analysis for Improved ReLU Network Feature Learning. arXiv preprint arXiv:2305.15912.

PDF Cite arXiv

Wenlin Chen, Austin Tripp, José Miguel Hernández-Lobato (2023). Meta-learning Adaptive Deep Kernel Gaussian Processes for Molecular Property Prediction. ICLR 2023.

PDF Cite Code OpenReview arXiv

Wenlin Chen, Samuel Horváth, Peter Richtárik (2022). Optimal Client Sampling for Federated Learning. TMLR 2022.

PDF Cite Code Video OpenReview arXiv

Austin Tripp, Wenlin Chen, José Miguel Hernández-Lobato (2022). An Evaluation Framework for the Objective Functions of De Novo Drug Design Benchmarks. ICLR 2022 MLDD Workshop.

PDF Cite OpenReview

Wenlin Chen (2021). Causal Representation Learning for Latent Space Optimization. MPhil Thesis, University of Cambridge.

PDF Cite

Andrew Webb, Charles Reynolds, Wenlin Chen, Henry Reeve, Dan Iliescu, Mikel Luján, Gavin Brown (2020). To Ensemble or Not Ensemble: When Does End-to-End Training Fail?. ECML 2020.

PDF Cite Code Video Springer Link arXiv