arXiv:2511.21054v1 Announce Type: new 
Abstract: Diffusion planning is a promising method for learning high-performance policies from offline data. To avoid the impact of discrepancies between planning and reality on performance, previous works generate new plans at each time step. However, this incurs significant computational overhead and leads to lower decision frequencies, and frequent plan switching may also affect performance. In contrast, humans might create detailed short-term plans and more general, sometimes vague, long-term plans, and adjust them over time. Inspired by this, we propose the Temporal Diffusion Planner (TDP) which improves decision efficiency by distributing the denoising steps across the time dimension. TDP begins by generating an initial plan that becomes progressively more vague over time. At each subsequent time step, rather than generating an entirely new plan, TDP updates the previous one with a small number of denoising steps. This reduces the average number of denoising steps, improving decision efficiency. Additionally, we introduce an automated replanning mechanism to prevent significant deviations between the plan and reality. Experiments on D4RL show that, compared to previous works that generate new plans every time step, TDP improves the decision-making frequency by 11-24.8 times while achieving higher or comparable performance.

تم اقتراح مخطط الانتشار الزمني (TDP) كنهج مبتكر للتخطيط الانتشاري، والذي يهدف إلى تحسين كفاءة اتخاذ القرار من خلال توزيع خطوات إزالة الضوضاء عبر الزمن. على عكس الطرق السابقة التي تولد خططًا جديدة في كل خطوة زمنية، يقوم TDP بتحديث الخطط السابقة، مما يسمح بعملية تخطيط أكثر تدريجية وكفاءة.

Se ha propuesto el Temporal Diffusion Planner (TDP) como un enfoque innovador para la planificación de difusión, que busca mejorar la eficiencia en la toma de decisiones al distribuir los pasos de desruido a lo largo del tiempo. A diferencia de los métodos anteriores que generan nuevos planes en cada paso, el TDP actualiza los planes existentes, permitiendo un proceso de planificación más gradual y eficiente.

Le Temporal Diffusion Planner (TDP) a été proposé comme une approche innovante de la planification de diffusion, visant à améliorer l'efficacité de la prise de décision en répartissant les étapes de débruitage dans le temps. Contrairement aux méthodes précédentes qui génèrent de nouveaux plans à chaque étape, le TDP met à jour les plans existants, permettant un processus de planification plus graduel et efficace.

The Temporal Diffusion Planner (TDP) has been proposed as an innovative approach to diffusion planning, which aims to enhance decision-making efficiency by distributing denoising steps over time. Unlike previous methods that generate new plans at every time step, TDP updates existing plans, allowing for a more gradual and efficient planning process.

Efficient Diffusion Planning with Temporal Diffusion

Was this article worth reading? Share it

LucidQuery AI

Metaflow AI

Routerra

Retentional's Success Planner

Agile Luminary

Solvice

Ready to build your own newsroom?