Timezone: »

A Pareto-optimal compositional energy-based model for sampling and optimization of protein sequences
Nataša Tagasovska · Nathan Frey · Andreas Loukas · Isidro Hotzel · Julien Lafrance-Vanasse · Ryan Kelly · Yan Wu · Arvind Rajpal · Richard Bonneau · Kyunghyun Cho · Stephen Ra · Vladimir Gligorijevic
Event URL: https://openreview.net/forum?id=U2rNXaTTXPQ »

Deep generative models have emerged as a popular machine learning-based approach for inverse design problems in the life sciences. However, these problems often require sampling new designs that satisfy multiple properties of interest in addition to learning the data distribution. This multi-objective optimization becomes more challenging when properties are independent or orthogonal to each other.In this work, we propose a Pareto-compositional energy-based model (pcEBM), a framework that uses multiple gradient descent for sampling new designs that adhere to various constraints in optimizing distinct properties. We demonstrate its ability to learn non-convex Pareto fronts and generate sequences that simultaneously satisfy multiple desired properties across a series of real-world antibody design tasks.

Author Information

Nataša Tagasovska (Prescient Design, Genentech)
Nathan Frey (Prescient Design • Genentech)
Andreas Loukas (Prescient Design, gRED, Roche)
Isidro Hotzel
Julien Lafrance-Vanasse (Genentech)
Ryan Kelly (Genentech)
Yan Wu
Arvind Rajpal
Richard Bonneau (New York University)
Kyunghyun Cho (Genentech / NYU)

Kyunghyun Cho is an associate professor of computer science and data science at New York University and a research scientist at Facebook AI Research. He was a postdoctoral fellow at the Université de Montréal until summer 2015 under the supervision of Prof. Yoshua Bengio, and received PhD and MSc degrees from Aalto University early 2014 under the supervision of Prof. Juha Karhunen, Dr. Tapani Raiko and Dr. Alexander Ilin. He tries his best to find a balance among machine learning, natural language processing, and life, but almost always fails to do so.

Stephen Ra (Prescient Design / Genentech)
Vladimir Gligorijevic (Prescient Design/Genentech)

More from the Same Authors