Learning Energy Decompositions for Partial Inference of GFlowNets

• URL: http://arxiv.org/abs/2310.03301v1
• Date: Thu, 5 Oct 2023 04:02:36 GMT
• Title: Learning Energy Decompositions for Partial Inference of GFlowNets
• Authors: Hyosoon Jang, Minsu Kim, Sungsoo Ahn
• Abstract summary: We study generative flow networks (GFlowNets) to sample objects from the Boltzmann energy distribution via a sequence of actions. In particular, we focus on improving GFlowNet with partial inference: training flow functions with the evaluation of the intermediate states or transitions.
• Score: 34.209530834968206
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper studies generative flow networks (GFlowNets) to sample objects from the Boltzmann energy distribution via a sequence of actions. In particular, we focus on improving GFlowNet with partial inference: training flow functions with the evaluation of the intermediate states or transitions. To this end, the recently developed forward-looking GFlowNet reparameterizes the flow functions based on evaluating the energy of intermediate states. However, such an evaluation of intermediate energies may (i) be too expensive or impossible to evaluate and (ii) even provide misleading training signals under large energy fluctuations along the sequence of actions. To resolve this issue, we propose learning energy decompositions for GFlowNets (LED-GFN). Our main idea is to (i) decompose the energy of an object into learnable potential functions defined on state transitions and (ii) reparameterize the flow functions using the potential functions. In particular, to produce informative local credits, we propose to regularize the potential to change smoothly over the sequence of actions. It is also noteworthy that training GFlowNet with our learned potential can preserve the optimal policy. We empirically verify the superiority of LED-GFN in five problems including the generation of unstructured and maximum independent sets, molecular graphs, and RNA sequences.

Related papers

• Pessimistic Backward Policy for GFlowNets [40.00805723326561]
  We study Generative Flow Networks (GFlowNets), which learn to sample objects proportionally to a given reward function. In this work, we observe that GFlowNets tend to under-exploit the high-reward objects due to training on insufficient number of trajectories. We propose a pessimistic backward policy for GFlowNets, which maximizes the observed flow to align closely with the true reward for the object.
  arXiv  Detail & Related papers  (2024-05-25T02:30:46Z)
• Pre-Training and Fine-Tuning Generative Flow Networks [61.90529626590415]
  We introduce a novel approach for reward-free pre-training of GFlowNets. By framing the training as a self-supervised problem, we propose an outcome-conditioned GFlowNet that learns to explore the candidate space. We show that the pre-trained OC-GFN model can allow for a direct extraction of a policy capable of sampling from any new reward functions in downstream tasks.
  arXiv  Detail & Related papers  (2023-10-05T09:53:22Z)
• Learning to Scale Logits for Temperature-Conditional GFlowNets [77.36931187299896]
  We propose textitLogit-scaling GFlowNets (Logit-GFN), a novel architectural design that greatly accelerates the training of temperature-conditional GFlowNets. We find that the challenge is greatly reduced if a learned function of the temperature is used to scale the policy's logits directly.
  arXiv  Detail & Related papers  (2023-10-04T13:45:56Z)
• Stochastic Generative Flow Networks [89.34644133901647]
  Generative Flow Networks (or GFlowNets) learn to sample complex structures through the lens of "inference as control" Existing GFlowNets can be applied only to deterministic environments, and fail in more general tasks with dynamics. This paper introduces GFlowNets, a new algorithm that extends GFlowNets to environments.
  arXiv  Detail & Related papers  (2023-02-19T03:19:40Z)
• Distributional GFlowNets with Quantile Flows [73.73721901056662]
  Generative Flow Networks (GFlowNets) are a new family of probabilistic samplers where an agent learns a policy for generating complex structure through a series of decision-making steps. In this work, we adopt a distributional paradigm for GFlowNets, turning each flow function into a distribution, thus providing more informative learning signals during training. Our proposed textitquantile matching GFlowNet learning algorithm is able to learn a risk-sensitive policy, an essential component for handling scenarios with risk uncertainty.
  arXiv  Detail & Related papers  (2023-02-11T22:06:17Z)
• Learning GFlowNets from partial episodes for improved convergence and stability [56.99229746004125]
  Generative flow networks (GFlowNets) are algorithms for training a sequential sampler of discrete objects under an unnormalized target density. Existing training objectives for GFlowNets are either local to states or transitions, or propagate a reward signal over an entire sampling trajectory. Inspired by the TD($lambda$) algorithm in reinforcement learning, we introduce subtrajectory balance or SubTB($lambda$), a GFlowNet training objective that can learn from partial action subsequences of varying lengths.
  arXiv  Detail & Related papers  (2022-09-26T15:44:24Z)
• Generative Flow Networks for Discrete Probabilistic Modeling [118.81967600750428]
  We present energy-based generative flow networks (EB-GFN) EB-GFN is a novel probabilistic modeling algorithm for high-dimensional discrete data. We show how GFlowNets can approximately perform large-block Gibbs sampling to mix between modes.
  arXiv  Detail & Related papers  (2022-02-03T01:27:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.