Variational Proximal Policy Optimization

• URL: http://arxiv.org/abs/2606.08032v1
• Date: Sat, 06 Jun 2026 07:50:50 GMT
• Title: Variational Proximal Policy Optimization
• Abstract summary: (textscVPtextscO) introduces a geometry-based proximal-control mechanism that can reduce reliance on fixed clipping or KL schedules.<n>(mathbf179) ELO gain on Codeforces and a (mathbf32%) reduction in token count on AIME mathematical reasoning tasks.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement Learning from Human Feedback via Proximal Policy Optimization often suffers from policy mode collapse, brittle exploration loops, and distribution drift. This paper introduces Variational Proximal Policy Optimization (\(\textsc{VP}_2\textsc{O}\)), a particle-based variational inference framework that maps policy optimization to Stein Variational Gradient Descent within a Mixture-of-Experts architecture. By leveraging functional kernels over localized expert prototypes alongside an expert orthogonalization loss, \(\textsc{VP}_2\textsc{O}\) introduces a geometry-based proximal-control mechanism that can reduce reliance on fixed clipping or KL schedules. Our results on a 33B/4B sparse Mixture-of-Experts model show several improvements across complex reasoning benchmarks, establishing a \(+\mathbf{179}\) ELO gain on Codeforces and a \(\mathbf{32\%}\) reduction in token count on AIME mathematical reasoning tasks.

Related papers

• BOOOM: Loss-Function-Agnostic Black-Box Optimization over Orthonormal Manifolds for Machine Learning and Statistical Inference [0.0]
  We present a systematic exploration of.<n>the set of $p.<n>orthonormal column-orthonormal escape through a.<n>novel applied to data data.<n>problems.<n>We show a diagonal optimization framework showing a.<n>strong, low-rank decomposition, independent decomposition analysis,.<n>and global convergence in under conditions.
  arXiv  Detail & Related papers  (2026-04-23T19:32:27Z)
• Orthogonalized Policy Optimization:Decoupling Sampling Geometry from Optimization Geometry in RLHF [0.0]
  Large language model alignment objectives are often presented as a collection of distinct algorithms, such as PPO, DPO, IPO, and their variants.<n>In this work, we argue that this diversity obscures a simpler underlying structure.<n>We show that this entanglement is not merely a modeling convenience but a source of systematic instability.
  arXiv  Detail & Related papers  (2026-01-18T13:57:44Z)
• A Kernel Approach for Semi-implicit Variational Inference [21.789560144560127]
  Semi-implicit variational inference (SIVI) enhances the expressiveness of variational families through hierarchical semi-implicit distributions.<n>Recent score-matching approaches to SIVI (SIVI-SM) address this issue via a minimax formulation.<n>We propose kernel semi-implicit variational inference (KSIVI), a principled and tractable alternative.
  arXiv  Detail & Related papers  (2026-01-17T12:06:12Z)
• Anisotropic Tensor Deconvolution of Hyperspectral Images [24.509592365283208]
  Hyperspectral image (HSI) deconvolution is a challenging ill-posed inverse problem, made difficult by the data's high dimensionality.<n>We propose a framework based on a low-parsimonious parameters of Canonical Poly Decomposition (CPD)<n>This approach recasts the problem from recovering a large-scale image with a latent to estimating the factors with $(P+Q+N)R$ variables.
  arXiv  Detail & Related papers  (2026-01-16T16:29:13Z)
• Graph Neural Regularizers for PDE Inverse Problems [62.49743146797144]
  We present a framework for solving a broad class of ill-posed inverse problems governed by partial differential equations (PDEs)<n>The forward problem is numerically solved using the finite element method (FEM)<n>We employ physics-inspired graph neural networks as learned regularizers, providing a robust, interpretable, and generalizable alternative to standard approaches.
  arXiv  Detail & Related papers  (2025-10-23T21:43:25Z)
• Neural Optimal Transport Meets Multivariate Conformal Prediction [58.43397908730771]
  We propose a framework for conditional vectorile regression (CVQR)<n>CVQR combines neural optimal transport with quantized optimization, and apply it to predictions.
  arXiv  Detail & Related papers  (2025-09-29T19:50:19Z)
• Multi-Sample Anti-Aliasing and Constrained Optimization for 3D Gaussian Splatting [6.336372495476242]
  We propose a comprehensive optimization framework integrating multisample anti-aliasing with dual geometric constraints.<n>Our system computes pixel colors through adaptive blending of quadruple subsamples, effectively reducing aliasing artifacts in high-frequency components.<n>Our method achieves state-of-the-art performance in detail preservation, particularly in preserving high-frequency textures and sharp discontinuities.
  arXiv  Detail & Related papers  (2025-08-14T10:14:36Z)
• Generalization Bounds of Surrogate Policies for Combinatorial Optimization Problems [53.03951222945921]
  We analyze smoothed (perturbed) policies, adding controlled random perturbations to the direction used by the linear oracle.<n>Our main contribution is a generalization bound that decomposes the excess risk into perturbation bias, statistical estimation error, and optimization error.<n>We illustrate the scope of the results on applications such as vehicle scheduling, highlighting how smoothing enables both tractable training and controlled generalization.
  arXiv  Detail & Related papers  (2024-07-24T12:00:30Z)
• Differentially Private Optimization with Sparse Gradients [60.853074897282625]
  We study differentially private (DP) optimization problems under sparsity of individual gradients. Building on this, we obtain pure- and approximate-DP algorithms with almost optimal rates for convex optimization with sparse gradients.
  arXiv  Detail & Related papers  (2024-04-16T20:01:10Z)
• Model-Based Reparameterization Policy Gradient Methods: Theory and Practical Algorithms [88.74308282658133]
  Reization (RP) Policy Gradient Methods (PGMs) have been widely adopted for continuous control tasks in robotics and computer graphics. Recent studies have revealed that, when applied to long-term reinforcement learning problems, model-based RP PGMs may experience chaotic and non-smooth optimization landscapes. We propose a spectral normalization method to mitigate the exploding variance issue caused by long model unrolls.
  arXiv  Detail & Related papers  (2023-10-30T18:43:21Z)
• Stable Nonconvex-Nonconcave Training via Linear Interpolation [51.668052890249726]
  This paper presents a theoretical analysis of linearahead as a principled method for stabilizing (large-scale) neural network training. We argue that instabilities in the optimization process are often caused by the nonmonotonicity of the loss landscape and show how linear can help by leveraging the theory of nonexpansive operators.
  arXiv  Detail & Related papers  (2023-10-20T12:45:12Z)
• PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback [106.63518036538163]
  We present a novel unified bilevel optimization-based framework, textsfPARL, formulated to address the recently highlighted critical issue of policy alignment in reinforcement learning. Our framework addressed these concerns by explicitly parameterizing the distribution of the upper alignment objective (reward design) by the lower optimal variable. Our empirical results substantiate that the proposed textsfPARL can address the alignment concerns in RL by showing significant improvements.
  arXiv  Detail & Related papers  (2023-08-03T18:03:44Z)
• Optimization on manifolds: A symplectic approach [127.54402681305629]
  We propose a dissipative extension of Dirac's theory of constrained Hamiltonian systems as a general framework for solving optimization problems. Our class of (accelerated) algorithms are not only simple and efficient but also applicable to a broad range of contexts.
  arXiv  Detail & Related papers  (2021-07-23T13:43:34Z)
• Adaptive Online Estimation of Piecewise Polynomial Trends [23.91519151164528]
  We consider the framework of non-stationary optimization with squared error losses and noisy gradient feedback. Motivated from the theory of non-parametric regression, we introduce a new variational constraint. We show that the same policy is minimax optimal for several other non-parametric families of interest.
  arXiv  Detail & Related papers  (2020-09-30T19:30:28Z)

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.