SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.05885 · ICML · 2024

GFlowNet Training by Policy Gradients

Xiaoning Qian, Puhua Niu, Shili Wu, Mingzhou Fan

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 4 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
niupuhua1234/gfn-pg canonical 0 of 4
FunctionStatusWhere it lives
download Not yet run niupuhua1234/gfn-pg/gfn-pg/data_dag/data.py
pointer only (licence: NONE) · get_code("874cb43146e2c04f")
ix_ Not yet run niupuhua1234/gfn-pg/gfn-pg/data_dag/scores/base.py
pointer only (licence: NONE) · get_code("a2b19bef0331f930")
logdet Not yet run niupuhua1234/gfn-pg/gfn-pg/data_dag/scores/base.py
pointer only (licence: NONE) · get_code("0cb1419ec492cafd")
logdet Not yet run niupuhua1234/gfn-pg/gfn-pg/data_dag/scores/base_torch.py
pointer only (licence: NONE) · get_code("52ac2620e1dc4258")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Generative Flow Networks (GFlowNets) have been shown effective to generate combinatorial objects with desired properties. We here propose a new GFlowNet training framework, with policydependent rewards, that bridges keeping flow balance of GFlowNets to optimizing the expected accumulated reward in traditional Reinforcement-Learning (RL). This enables the derivation of new policy-based GFlowNet training methods, in contrast to existing ones resembling value-based RL. It is known that the design of backward policies in GFlowNet training affects efficiency. We further develop a coupled training strategy that jointly solves GFlowNet forward policy training and backward policy design. Performance analysis is provided with a theoretical guarantee of our policy-based GFlowNet training. Experiments on both simulated and real-world datasets verify that our policy-based strategies provide advanced RL perspectives for robust gradient estimation to improve GFlowNet performance. Our code is

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2408.05885")
get_code_for_paper("2408.05885")
have("2408.05885")

Connect an agent — have() is free.