Timezone: »

Towards Understanding and Improving GFlowNet Training
Max Shen · Emmanuel Bengio · Ehsan Hajiramezanali · Andreas Loukas · Kyunghyun Cho · Tommaso Biancalani

Wed Jul 26 02:00 PM -- 03:30 PM (PDT) @ Exhibit Hall 1 #535
Generative flow networks (GFlowNets) are a family of algorithms that learn a generative policy to sample discrete objects $x$ with non-negative reward $R(x)$. Learning objectives guarantee the GFlowNet samples $x$ from the target distribution $p^*(x) \propto R(x)$ when loss is globally minimized over all states or trajectories, but it is unclear how well they perform with practical limits on training resources. We introduce an efficient evaluation strategy to compare the learned sampling distribution to the target reward distribution. As flows can be underdetermined given training data, we clarify the importance of learned flows to generalization and matching $p^*(x)$ in practice. We investigate how to learn better flows, and propose (i) prioritized replay training of high-reward $x$, (ii) relative edge flow policy parametrization, and (iii) a novel guided trajectory balance objective, and show how it can solve a substructure credit assignment problem. We substantially improve sample efficiency on biochemical design tasks.

Author Information

Max Shen (Genentech)
Emmanuel Bengio (McGill University)
Ehsan Hajiramezanali (Genentech)
Andreas Loukas (EPFL)
Kyunghyun Cho (New York University, Genentech)
Kyunghyun Cho

Kyunghyun Cho is an associate professor of computer science and data science at New York University and CIFAR Fellow of Learning in Machines & Brains. He is also a senior director of frontier research at the Prescient Design team within Genentech Research & Early Development (gRED). He was a research scientist at Facebook AI Research from June 2017 to May 2020 and a postdoctoral fellow at University of Montreal until Summer 2015 under the supervision of Prof. Yoshua Bengio, after receiving MSc and PhD degrees from Aalto University April 2011 and April 2014, respectively, under the supervision of Prof. Juha Karhunen, Dr. Tapani Raiko and Dr. Alexander Ilin. He received the Samsung Ho-Am Prize in Engineering in 2021. He tries his best to find a balance among machine learning, natural language processing, and life, but almost always fails to do so.

Tommaso Biancalani (Genentech)
Tommaso Biancalani

I run the department of Machine Learning for Biology in Genentech

More from the Same Authors