Avatar

Alvin Chan

Postdoctoral Associate

Massachusetts Institute of Technology (MIT), USA

Biography

Alvin Chan is a postdoctoral associate in Prof Giovanni Traverso's team, at MIT, to work on designing medical nanoparticles with deep learning. His research interests center around making AI technologies beneficial and safe for humanity, encompassing topics like generative modeling, adversarial learning and deep learning for medical applications. In his free time, he enjoys rock-climbing, jogging and attending tech meetups.

Interests

  • Generative Models
  • Computational Biology
  • Adversarial Learning

Education

  • PhD in Computer Science, 2021

    Nanyang Technological University, Singapore

  • BEng in Bioengineering, 2013

    Nanyang Technological University, Singapore

Recent Publications

Deep Extrapolation for Attribute-Enhanced Generation

TL;DR: How do we generate sequences that extrapolate beyond the training distribution?
Abstract: Attribute extrapolation in sample …

Self-Instantiated Recurrent Units with Dynamic Soft Recursion

TL;DR: We propose the self-instantiated recurrent unit that is characterized by recursive instantiation of the model itself, where the …

On Orthogonality Constraints for Transformers

TL;DR: Orthogonality constraints the encourages numerical stability improves model's performance in NLP tasks.
Abstract: Orthogonality …

CoCon: A Self-Supervised Approach for Controlled Text Generation

TL;DR: We propose CoCon to control the content of text generation from LMs by conditioning on content inputs at an interleave layer. …

Parameterization of Hypercomplex Multiplications

TL;DR: We propose a new parameterization of hypercomplex multiplications for architectural flexibility and effectiveness.
Abstract: …