arXiv:2512.02323v1 Announce Type: cross 
Abstract: Boltzmann machines (BMs) are powerful energy-based generative models, but their heavy training cost has largely confined practical use to Restricted BMs (RBMs) trained with an efficient learning method called contrastive divergence. More accurate learning typically requires Markov chain Monte Carlo (MCMC) Boltzmann sampling, but it is time-consuming due to the difficulty of parallelization for more expressive models. To address this limitation, we first propose a new Boltzmann sampler inspired by a quantum-inspired combinatorial optimization called simulated bifurcation (SB). This SB-inspired approach, which we name Langevin SB (LSB), enables parallelized sampling while maintaining accuracy comparable to MCMC. Furthermore, this is applicable not only to RBMs but also to BMs with general couplings. However, LSB cannot control the inverse temperature of the output Boltzmann distribution, which hinders learning and degrades performance. To overcome this limitation, we also developed an efficient method for estimating the inverse temperature during the learning process, which we call conditional expectation matching (CEM). By combining LSB and CEM, we establish an efficient learning framework for BMs with greater expressive power than RBMs. We refer to this framework as sampler-adaptive learning (SAL). SAL opens new avenues for energy-based generative modeling beyond RBMs.

تم اقتراح نهج جديد لآلات بولتزمان (BM) ، حيث تم تقديم عينة قابلة للتوازي تُسمى Langevin SB (LSB) ، مستوحاة من الانقسام المحاكي. تُحسن هذه الطريقة من كفاءة العينة مع الحفاظ على دقة مقارنة بتقنيات مونت كارلو التقليدية عبر سلاسل ماركوف ، مما يوسع من قابلية تطبيق آلات بولتزمان إلى ما هو أبعد من آلات بولتزمان المقيدة (RBM).

Se ha propuesto un nuevo enfoque para las máquinas de Boltzmann (BM), introduciendo un muestreador paralelizable llamado Langevin SB (LSB), inspirado en la bifurcación simulada. Este método mejora la eficiencia del muestreo mientras mantiene una precisión comparable a las técnicas tradicionales de Monte Carlo por cadenas de Markov, ampliando la aplicabilidad de las BM más allá de las BM restringidas (RBM).

Une nouvelle approche des machines de Boltzmann (BM) a été proposée, introduisant un échantillonneur parallélisable appelé Langevin SB (LSB), inspiré par la bifurcation simulée. Cette méthode améliore l'efficacité de l'échantillonnage tout en maintenant une précision comparable aux techniques traditionnelles de Monte Carlo par chaînes de Markov, élargissant l'applicabilité des BM au-delà des BM restreintes (RBM).

A new approach to Boltzmann machines (BMs) has been proposed, introducing a parallelizable sampler called Langevin SB (LSB), inspired by simulated bifurcation. This method enhances sampling efficiency while maintaining accuracy comparable to traditional Markov chain Monte Carlo techniques, expanding the applicability of BMs beyond Restricted BMs (RBMs).

Unlocking the Power of Boltzmann Machines by Parallelizable Sampler and Efficient Temperature Estimation

arXiv:2508.18175v2 Announce Type: replace 
Abstract: Efficient equilibrium sampling of molecular conformations remains a core challenge in computational chemistry and statistical inference. Classical approaches such as molecular dynamics or Markov chain Monte Carlo inherently lack amortization; the computational cost of sampling must be paid in full for each system of interest. The widespread success of generative models has inspired interest towards overcoming this limitation through learning sampling algorithms. Despite performing competitively with conventional methods when trained on a single system, learned samplers have so far demonstrated limited ability to transfer across systems. We demonstrate that deep learning enables the design of scalable and transferable samplers by introducing Prose, a 285 million parameter all-atom transferable normalizing flow trained on a corpus of peptide molecular dynamics trajectories up to 8 residues in length. Prose draws zero-shot uncorrelated proposal samples for arbitrary peptide systems, achieving the previously intractable transferability across sequence length, whilst retaining the efficient likelihood evaluation of normalizing flows. Through extensive empirical evaluation we demonstrate the efficacy of Prose as a proposal for a variety of sampling algorithms, finding a simple importance sampling-based finetuning procedure to achieve competitive performance to established methods such as sequential Monte Carlo. We open-source the Prose codebase, model weights, and training dataset, to further stimulate research into amortized sampling methods and finetuning objectives.

تقدم دراسة جديدة Prose، وهو تدفق طبيعي قابل للنقل يتكون من 285 مليون معلمة مصمم لعينات فعالة من التكوينات الجزيئية، متجاوزًا القيود المفروضة على الأساليب التقليدية مثل الديناميكا الجزيئية وMonte Carlo بواسطة السلاسل. تتيح هذه التطورات أخذ عينات غير مرتبطة في صفر ضربة عبر أنظمة جزيئية مختلفة، مما يعزز الكفاءة الحسابية في الكيمياء.

Un nuevo estudio presenta Prose, un flujo normalizador transferible de 285 millones de parámetros diseñado para un muestreo eficiente de conformaciones moleculares, superando las limitaciones de métodos tradicionales como la dinámica molecular y el muestreo de Monte Carlo por cadenas. Este avance permite un muestreo no correlacionado en cero disparos a través de diferentes sistemas moleculares, mejorando la eficiencia computacional en química.

Une nouvelle étude présente Prose, un flux normalisant transférable de 285 millions de paramètres conçu pour un échantillonnage efficace des conformations moléculaires, surmontant les limitations des méthodes traditionnelles telles que la dynamique moléculaire et le Monte Carlo par chaînes de Markov. Cette avancée permet un échantillonnage non corrélé en zéro coup à travers différents systèmes moléculaires, améliorant l'efficacité computationnelle en chimie.

A new study introduces Prose, a 285 million parameter transferable normalizing flow designed for efficient sampling of molecular conformations, overcoming limitations of traditional methods like molecular dynamics and Markov chain Monte Carlo. This advancement allows for zero-shot uncorrelated sampling across different molecular systems, enhancing computational efficiency in chemistry.

Unlocking the Power of Boltzmann Machines by Parallelizable Sampler and Efficient Temperature Estimation

Was this article worth reading? Share it

LucidQuery AI

Airparser

FastML