Generative Flow Networks for Discrete Probabilistic Modeling

2022/02/03 by Dinghuai Zhang, Nikolay Malkin, Zhang, Dinghuai +9 · 10 citations
Computer Science · #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Human Pose and Action Recognition #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Neural Networks and Applications

paper · pdf · doi:10.48550/arxiv.2202.01361

openalex publication_date 2022/02/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present energy-based generative flow networks (EB-GFN), a novel probabilistic modeling algorithm for high-dimensional discrete data. Building upon the theory of generative flow networks (GFlowNets), we model the generation process by a stochastic data construction policy and thus amortize expensive MCMC exploration into a fixed number of actions sampled from a GFlowNet. We show how GFlowNets can approximately perform large-block Gibbs sampling to mix between modes. We propose a framework to jointly train a GFlowNet with an energy function, so that the GFlowNet learns to sample from the energy distribution, while the energy learns with an approximate MLE objective with negative samples from the GFlowNet. We demonstrate EB-GFN's effectiveness on various probabilistic modeling tasks. Code is publicly available at https://github.com/zdhNarsil/EB_GFN.

Cited by

• Value Gradient Guidance for Flow Matching Alignment
• gfnx: Fast and Scalable Library for Generative Flow Networks in JAX
• Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need?
• Latent Chain-of-Thought for Visual Reasoning
• Reinforced sequential Monte Carlo for amortised sampling
• MG2FlowNet: Accelerating High-Reward Sample Generation via Enhanced MCTS and Greediness Control
• FlowRL: Matching Reward Distributions for LLM Reasoning
• Time-Modulated Intelligent Reflecting Surfaces for Integrated Sensing, Communication and Security: A Generative AI Design Framework
• Mitigating Hallucinations in LM-Based TTS Models via Distribution Alignment Using GFlowNets
• Expert-aided causal discovery of ancestral graphs

Related

• Stochastic Generative Flow Networks
• Generative Flow Networks: a Markov Chain Perspective
• GFlowNet Foundations
• Bifurcated Generative Flow Networks
• Investigating Generalization Behaviours of Generative Flow Networks