Control, Transport and Sampling: Towards Better Loss Design

• URL: http://arxiv.org/abs/2405.13731v2
• Date: Thu, 10 Oct 2024 17:16:30 GMT
• Title: Control, Transport and Sampling: Towards Better Loss Design
• Authors: Qijia Jiang, David Nabergoj,
• Abstract summary: We propose objective functions that can be used to transport $nu$ to $mu$, via optimally controlled dynamics. We highlight the importance of the pathwise perspective and the role various optimality conditions on the path measure can play for the design of valid training losses.
• Score: 10.732151772173083
• License:
• Abstract: Leveraging connections between diffusion-based sampling, optimal transport, and stochastic optimal control through their shared links to the Schr\"odinger bridge problem, we propose novel objective functions that can be used to transport $\nu$ to $\mu$, consequently sample from the target $\mu$, via optimally controlled dynamics. We highlight the importance of the pathwise perspective and the role various optimality conditions on the path measure can play for the design of valid training losses, the careful choice of which offer numerical advantages in implementation. Basing the formalism on Schr\"odinger bridge comes with the additional practical capability of baking in inductive bias when it comes to Neural Network training.

Related papers

• Feedback Schrödinger Bridge Matching [21.5490272744586]
  We introduce Feedback Schr"odinger Bridge Matching (FSBM), a novel semi-supervised matching framework. FSBM incorporates a small portion (less than 8% of the entire dataset) of pre-aligned pairs as state feedback to guide the transport map of non coupled samples. Extensive experiments demonstrate that FSBM accelerates training and enhances generalization by leveraging coupled pairs guidance.
  arXiv  Detail & Related papers  (2024-10-17T21:52:01Z)
• BM$^2$: Coupled Schrödinger Bridge Matching [4.831663144935879]
  We introduce a simple emphnon-iterative approach for learning Schr"odinger bridges with neural networks. A preliminary theoretical analysis of the convergence properties of BM$2$ is carried out, supported by numerical experiments.
  arXiv  Detail & Related papers  (2024-09-14T08:57:46Z)
• Plug-in estimation of Schrödinger bridges [15.685006881635209]
  We propose a procedure for estimating the Schr"odinger bridge between two probability distributions. We show that our proposal, which we call the emphSinkhorn bridge, provably estimates the Schr"odinger bridge with a rate of convergence that depends on the intrinsic dimensionality of the target measure.
  arXiv  Detail & Related papers  (2024-08-21T15:07:25Z)
• Dynamical Measure Transport and Neural PDE Solvers for Sampling [77.38204731939273]
  We tackle the task of sampling from a probability density as transporting a tractable density function to the target. We employ physics-informed neural networks (PINNs) to approximate the respective partial differential equations (PDEs) solutions. PINNs allow for simulation- and discretization-free optimization and can be trained very efficiently.
  arXiv  Detail & Related papers  (2024-07-10T17:39:50Z)
• Hallmarks of Optimization Trajectories in Neural Networks: Directional Exploration and Redundancy [75.15685966213832]
  We analyze the rich directional structure of optimization trajectories represented by their pointwise parameters. We show that training only scalar batchnorm parameters some while into training matches the performance of training the entire network.
  arXiv  Detail & Related papers  (2024-03-12T07:32:47Z)
• A Computational Framework for Solving Wasserstein Lagrangian Flows [48.87656245464521]
  In general, the optimal density path is unknown, and solving these variational problems can be computationally challenging. We propose a novel deep learning based framework approaching all of these problems from a unified perspective. We showcase the versatility of the proposed framework by outperforming previous approaches for the single-cell trajectory inference.
  arXiv  Detail & Related papers  (2023-10-16T17:59:54Z)
• Generalized Schrödinger Bridge Matching [54.171931505066]
  Generalized Schr"odinger Bridge (GSB) problem setup is prevalent in many scientific areas both within and without machine learning. We propose Generalized Schr"odinger Bridge Matching (GSBM), a new matching algorithm inspired by recent advances. We show that such a generalization can be cast as solving conditional optimal control, for which variational approximations can be used.
  arXiv  Detail & Related papers  (2023-10-03T17:42:11Z)
• Integrated Sensing, Computation, and Communication for UAV-assisted Federated Edge Learning [52.7230652428711]
  Federated edge learning (FEEL) enables privacy-preserving model training through periodic communication between edge devices and the server. Unmanned Aerial Vehicle (UAV)mounted edge devices are particularly advantageous for FEEL due to their flexibility and mobility in efficient data collection.
  arXiv  Detail & Related papers  (2023-06-05T16:01:33Z)
• Path Integral Sampler: a stochastic control approach for sampling [4.94950858749529]
  We present Path Integral Sampler(PIS), a novel algorithm to draw samples from unnormalized probability density functions. The PIS draws samples from the initial distribution and then propagates the samples through the Schr"odinger bridge to reach the terminal distribution. We provide theoretical justification of the sampling quality of PIS in terms of Wasserstein distance when sub-optimal control is used.
  arXiv  Detail & Related papers  (2021-11-30T05:50:12Z)
• Reinforcement Learning for Low-Thrust Trajectory Design of Interplanetary Missions [77.34726150561087]
  This paper investigates the use of reinforcement learning for the robust design of interplanetary trajectories in presence of severe disturbances. An open-source implementation of the state-of-the-art algorithm Proximal Policy Optimization is adopted. The resulting Guidance and Control Network provides both a robust nominal trajectory and the associated closed-loop guidance law.
  arXiv  Detail & Related papers  (2020-08-19T15:22:15Z)
• Large-Scale Optimal Transport via Adversarial Training with Cycle-Consistency [30.305690062622283]
  We propose an end-to-end approach for large-scale optimal transport, which directly solves the transport map and is compatible with general cost function. We demonstrate the effectiveness of the proposed method against existing methods with large-scale real-world applications.
  arXiv  Detail & Related papers  (2020-03-14T14:06:46Z)

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.