Search papers, labs, and topics across Lattice.
WINFlowNets is introduced as a novel CFlowNets framework that enables the co-training of flow and retrieval networks, removing the need for pre-training the retrieval network. It achieves this by incorporating a warm-up phase for the retrieval network, followed by a shared training architecture and replay buffer for co-training. Experiments in simulated robotic environments demonstrate that WINFlowNets outperforms CFlowNets and state-of-the-art RL algorithms in terms of average reward, training stability, and adaptation to fault environments.
By co-training flow and retrieval networks, WINFlowNets eliminates the need for pre-training, unlocking CFlowNets for dynamic robotic environments where data is scarce.
Generative Flow Networks for continuous scenarios (CFlowNets) have shown promise in solving sequential decision-making tasks by learning stochastic policies using a flow and a retrieval network. Despite their demonstrated efficiency compared to state-of-the-art Reinforcement Learning (RL) algorithms, their practical application in robotic control tasks is constrained by the reliance on pre-training the retrieval network. This dependency poses challenges in dynamic robotic environments, where pre-training data may not be readily available or representative of the current environment. This paper introduces WINFlowNets, a novel CFlowNets framework that enables the co-training of flow and retrieval networks. WINFlowNets begins with a warm-up phase for the retrieval network to bootstrap its policy, followed by a shared training architecture and a shared replay buffer for co-training both networks. Experiments in simulated robotic environments demonstrate that WINFlowNets surpasses CFlowNets and state-of-the-art RL algorithms in terms of average reward and training stability. Furthermore, WINFlowNets exhibits strong adaptive capability in fault environments, making it suitable for tasks that demand quick adaptation with limited sample data. These findings highlight WINFlowNets'potential for deployment in dynamic and malfunction-prone robotic systems, where traditional pre-training or sample inefficient data collection may be impractical.
