Proximal Oracles for Optimization and Sampling

We consider convex optimization with non-smooth objective function and log-concave sampling with non-smooth potential (negative log density). In particular, we study two specific settings where the convex objective/potential function is either semi-smooth or in composite form as the finite sum of semi-smooth components. To overcome the challenges caused by non-smoothness, our algorithms employ two powerful proximal frameworks in optimization and sampling: the proximal point framework for optimization and the alternating sampling framework (ASF) that uses Gibbs sampling on an augmented distribution. A key component of both optimization and sampling algorithms is the efficient implementation of the proximal map by the regularized cutting-plane method. We establish the iteration-complexity of the proximal map in both semi-smooth and composite settings. We further propose an adaptive proximal bundle method for non-smooth optimization. The proposed method is universal since it does not need any problem parameters as input. Additionally, we develop a proximal sampling oracle that resembles the proximal map in optimization and establish its complexity using a novel technique (a modified Gaussian integral). Finally, we combine this proximal sampling oracle and ASF to obtain a Markov chain Monte Carlo method with non-asymptotic complexity bounds for sampling in semi-smooth and composite settings.

Efficient Belief Road Map for Planning Under Uncertainty

Robotic systems, particularly in demanding environments like narrow corridors or disaster zones, often grapple with imperfect state estimation. Addressing this challenge requires a trajectory plan that not only navigates these restrictive spaces but also manages the inherent uncertainty of the system. We present a novel approach for graph-based belief space planning via the use of an efficient covariance control algorithm. By adaptively steering state statistics via output state feedback, we efficiently craft a belief roadmap characterized by nodes with controlled uncertainty and edges representing collision-free mean trajectories. The roadmap's structured design then paves the way for precise path searches that balance control costs and uncertainty considerations. Our numerical experiments affirm the efficacy and advantage of our method in different motion planning tasks. Our open-source implementation can be found at https://github.com/hzyu17/VIMP/tree/BRM.

On the Contraction Coefficient of the Schrödinger Bridge for Stochastic Linear Systems

Schr\"{o}dinger bridge is a stochastic optimal control problem to steer a given initial state density to another, subject to controlled diffusion and deadline constraints. A popular method to numerically solve the Schr\"{o}dinger bridge problems, in both classical and in the linear system settings, is via contractive fixed point recursions. These recursions can be seen as dynamic versions of the well-known Sinkhorn iterations, and under mild assumptions, they solve the so-called Schr\"{o}dinger systems with guaranteed linear convergence. In this work, we study a priori estimates for the contraction coefficients associated with the convergence of respective Schr\"{o}dinger systems. We provide new geometric and control-theoretic interpretations for the same. Building on these newfound interpretations, we point out the possibility of improved computation for the worst-case contraction coefficients of linear SBPs by preconditioning the endpoint support sets.

Stochastic Motion Planning as Gaussian Variational Inference: Theory and Algorithms

We consider the motion planning problem under uncertainty and address it using probabilistic inference. A collision-free motion plan with linear stochastic dynamics is modeled by a posterior distribution. Gaussian variational inference is an optimization over the path distributions to infer this posterior within the scope of Gaussian distributions. We propose Gaussian Variational Inference Motion Planner (GVI-MP) algorithm to solve this Gaussian inference, where a natural gradient paradigm is used to iteratively update the Gaussian distribution, and the factorized structure of the joint distribution is leveraged. We show that the direct optimization over the state distributions in GVI-MP is equivalent to solving a stochastic control that has a closed-form solution. Starting from this observation, we propose our second algorithm, Proximal Gradient Covariance Steering Motion Planner (PGCS-MP), to solve the same inference problem in its stochastic control form with terminal constraints. We use a proximal gradient paradigm to solve the linear stochastic control with nonlinear collision cost, where the nonlinear cost is iteratively approximated using quadratic functions and a closed-form solution can be obtained by solving a linear covariance steering at each iteration. We evaluate the effectiveness and the performance of the proposed approaches through extensive experiments on various robot models. The code for this paper can be found in https://github.com/hzyu17/VIMP.

Online Control for Linear Dynamics: A Data-Driven Approach

This paper considers an online control problem over a linear time-invariant system with unknown dynamics, bounded disturbance, and adversarial cost. We propose a data-driven strategy to reduce the regret of the controller. Unlike model-based methods, our algorithm does not identify the system model, instead, it leverages a single noise-free trajectory to calculate the accumulation of disturbance and makes decisions using the accumulated disturbance action controller we design, whose parameters are updated by online gradient descent. We prove that the regret of our algorithm is $\mathcal{O}(\sqrt{T})$ under mild assumptions, suggesting that its performance is on par with model-based methods.

Improved Order Analysis and Design of Exponential Integrator for Diffusion Models Sampling

Efficient differential equation solvers have significantly reduced the sampling time of diffusion models (DMs) while retaining high sampling quality. Among these solvers, exponential integrators (EI) have gained prominence by demonstrating state-of-the-art performance. However, existing high-order EI-based sampling algorithms rely on degenerate EI solvers, resulting in inferior error bounds and reduced accuracy in contrast to the theoretically anticipated results under optimal settings. This situation makes the sampling quality extremely vulnerable to seemingly innocuous design choices such as timestep schedules. For example, an inefficient timestep scheduler might necessitate twice the number of steps to achieve a quality comparable to that obtained through carefully optimized timesteps. To address this issue, we reevaluate the design of high-order differential solvers for DMs. Through a thorough order analysis, we reveal that the degeneration of existing high-order EI solvers can be attributed to the absence of essential order conditions. By reformulating the differential equations in DMs and capitalizing on the theory of exponential integrators, we propose refined EI solvers that fulfill all the order conditions, which we designate as Refined Exponential Solver (RES). Utilizing these improved solvers, RES exhibits more favorable error bounds theoretically and achieves superior sampling efficiency and stability in practical applications. For instance, a simple switch from the single-step DPM-Solver++ to our order-satisfied RES solver when Number of Function Evaluations (NFE) $=9$, results in a reduction of numerical defects by $25.2\%$ and FID improvement of $25.4\%$ (16.77 vs 12.51) on a pre-trained ImageNet diffusion model.

Diffusion-Based Adversarial Sample Generation for Improved Stealthiness and Controllability

Neural networks are known to be susceptible to adversarial samples: small variations of natural examples crafted to deliberately mislead the models. While they can be easily generated using gradient-based techniques in digital and physical scenarios, they often differ greatly from the actual data distribution of natural images, resulting in a trade-off between strength and stealthiness. In this paper, we propose a novel framework dubbed Diffusion-Based Projected Gradient Descent (Diff-PGD) for generating realistic adversarial samples. By exploiting a gradient guided by a diffusion model, Diff-PGD ensures that adversarial samples remain close to the original data distribution while maintaining their effectiveness. Moreover, our framework can be easily customized for specific tasks such as digital attacks, physical-world attacks, and style-based attacks. Compared with existing methods for generating natural-style adversarial samples, our framework enables the separation of optimizing adversarial loss from other surrogate losses (e.g., content/smoothness/style loss), making it more stable and controllable. Finally, we demonstrate that the samples generated using Diff-PGD have better transferability and anti-purification power than traditional gradient-based methods. Code will be released in https://github.com/xavihart/Diff-PGD

DiffCollage: Parallel Generation of Large Content with Diffusion Models

We present DiffCollage, a compositional diffusion model that can generate large content by leveraging diffusion models trained on generating pieces of the large content. Our approach is based on a factor graph representation where each factor node represents a portion of the content and a variable node represents their overlap. This representation allows us to aggregate intermediate outputs from diffusion models defined on individual nodes to generate content of arbitrary size and shape in parallel without resorting to an autoregressive generation procedure. We apply DiffCollage to various tasks, including infinite image generation, panorama image generation, and long-duration text-guided motion generation. Extensive experimental results with a comparison to strong autoregressive baselines verify the effectiveness of our approach.

Improved dimension dependence of a proximal algorithm for sampling

We propose a sampling algorithm that achieves superior complexity bounds in all the classical settings (strongly log-concave, log-concave, Logarithmic-Sobolev inequality (LSI), Poincar\'e inequality) as well as more general settings with semi-smooth or composite potentials. Our algorithm is based on the proximal sampler introduced in~\citet{lee2021structured}. The performance of this proximal sampler is determined by that of the restricted Gaussian oracle (RGO), a key step in the proximal sampler. The main contribution of this work is an inexact realization of RGO based on approximate rejection sampling. To bound the inexactness of RGO, we establish a new concentration inequality for semi-smooth functions over Gaussian distributions, extending the well-known concentration inequality for Lipschitz functions. Applying our RGO implementation to the proximal sampler, we achieve state-of-the-art complexity bounds in almost all settings. For instance, for strongly log-concave distributions, our method has complexity bound $\tilde\mathcal{O}(\kappa d^{1/2})$ without warm start, better than the minimax bound for MALA. For distributions satisfying the LSI, our bound is $\tilde \mathcal{O}(\hat \kappa d^{1/2})$ where $\hat \kappa$ is the ratio between smoothness and the LSI constant, better than all existing bounds.

A Gaussian variational inference approach to motion planning

We propose a Gaussian variational inference framework for the motion planning problem. In this framework, motion planning is formulated as an optimization over the distribution of the trajectories to approximate the desired trajectory distribution by a tractable Gaussian distribution. Equivalently, the proposed framework can be viewed as a standard motion planning with an entropy regularization. Thus, the solution obtained is a transition from an optimal deterministic solution to a stochastic one, and the proposed framework can recover the deterministic solution by controlling the level of stochasticity. To solve this optimization, we adopt the natural gradient descent scheme. The sparsity structure of the proposed formulation induced by factorized objective functions is further leveraged to improve the scalability of the algorithm. We evaluate our method on several robot systems in simulated environments, and show that it achieves collision avoidance with smooth trajectories, and meanwhile brings robustness to the deterministic baseline results, especially in challenging environments and tasks.