We study population convergence guarantees of stochastic gradient descent (SGD) for smooth convex objectives in the interpolation regime, where the noise at optimum is zero or near zero. The behavior of the last iterate of SGD in this setting -- particularly with large (constant) stepsizes -- has received growing attention in recent years due to implications for the training of over-parameterized models, as well as to analyzing forgetting in continual learning and to understanding the convergence of the randomized Kaczmarz method for solving linear systems. We establish that after $T$ steps of SGD on $β$-smooth convex loss functions with stepsize $0 < η< 2/β$, the last iterate exhibits expected excess risk $\widetilde{O}(\frac{1}{η(2-βη) T^{1-βη/2}} + \fracη{(2-βη)^2} T^{βη/2} σ_\star^2)$, where $σ_\star^2$ denotes the variance of the stochastic gradients at the optimum. In particular, for a well-tuned stepsize we obtain a near optimal $\widetilde{O}(1/T + σ_\star/\sqrt{T})$ rate for the last iterate, extending the results of Varre et al. (2021) beyond least squares regression; and when $σ_\star=0$ we obtain a rate of $\smash{O(1/\sqrt T)}$ with $η=1/β$, improving upon the best-known $\smash{O(T^{-1/4})}$ rate recently established by Evron et al. (2025) in the special case of realizable linear regression.

Graph-structured data is prevalent in many real-world applications, including social networks, financial systems, and molecular biology. Graph Neural Networks (GNNs) have become the de facto standard for learning from such data due to their strong representation capabilities. As GNNs are increasingly deployed in federated learning (FL) settings to preserve data locality and privacy, new privacy threats arise from the interaction between graph structures and decentralized training. In this paper, we present the first systematic study of cross-client membership inference attacks (CC-MIA) against node classification tasks of federated GNNs (FedGNNs), where a malicious client aims to infer which client owns the given data. Unlike prior centralized-focused work that focuses on whether a sample was included in training, our attack targets sample-to-client attribution, a finer-grained privacy risk unique to federated settings. We design a general attack framework that exploits FedGNNs' aggregation behaviors, gradient updates, and embedding proximity to link samples to their source clients across training rounds. We evaluate our attack across multiple graph datasets under realistic FL setups. Results show that our method achieves high performance on both membership inference and ownership identification. Our findings highlight a new privacy threat in federated graph learning-client identity leakage through structural and model-level cues, motivating the need for attribution-robust GNN design.

Transformer models have emerged as fundamental tools across various scientific and engineering disciplines, owing to their outstanding performance in diverse applications. Despite this empirical success, the theoretical foundations of Transformers remain relatively underdeveloped, particularly in understanding their training dynamics. Existing research predominantly examines isolated components--such as self-attention mechanisms and feedforward networks--without thoroughly investigating the interdependencies between these components, especially when residual connections are present. In this paper, we aim to bridge this gap by analyzing the convergence behavior of a structurally complete yet single-layer Transformer, comprising self-attention, a feedforward network, and residual connections. We demonstrate that, under appropriate initialization, gradient descent exhibits a linear convergence rate, where the convergence speed is determined by the minimum and maximum singular values of the output matrix from the attention layer. Moreover, our analysis reveals that residual connections serve to ameliorate the ill-conditioning of this output matrix, an issue stemming from the low-rank structure imposed by the softmax operation, thereby promoting enhanced optimization stability. We also extend our theoretical findings to a multi-layer Transformer architecture, confirming the linear convergence rate of gradient descent under suitable initialization. Empirical results corroborate our theoretical insights, illustrating the beneficial role of residual connections in promoting convergence stability.

Denoiser models have become powerful tools for inverse problems, enabling the use of pretrained networks to approximate the score of a smoothed prior distribution. These models are often used in heuristic iterative schemes aimed at solving Maximum a Posteriori (MAP) optimisation problems, where the proximal operator of the negative log-prior plays a central role. In practice, this operator is intractable, and practitioners plug in a pretrained denoiser as a surrogate-despite the lack of general theoretical justification for this substitution. In this work, we show that a simple algorithm, closely related to several used in practice, provably converges to the proximal operator under a log-concavity assumption on the prior $p$. We show that this algorithm can be interpreted as a gradient descent on smoothed proximal objectives. Our analysis thus provides a theoretical foundation for a class of empirically successful but previously heuristic methods.

As a paradigm for sequential decision making in unknown environments, reinforcement learning (RL) has received a flurry of attention in recent years. However, the explosion of model complexity in emerging applications and the presence of nonconvexity exacerbate the challenge of achieving efficient RL in sample-starved situations, where data collection is expensive, time-consuming, or even high-stakes (e.g., in clinical trials, autonomous systems, and online advertising). How to understand and enhance the sample and computational efficacies of RL algorithms is thus of great interest. In this tutorial, we aim to introduce several important algorithmic and theoretical developments in RL, highlighting the connections between new ideas and classical topics. Employing Markov Decision Processes as the central mathematical model, we cover several distinctive RL scenarios (i.e., RL with a simulator, online RL, offline RL, robust RL, and RL with human feedback), and present several mainstream RL approaches (i.e., model-based approach, value-based approach, and policy optimization). Our discussions gravitate around the issues of sample complexity, computational efficiency, as well as algorithm-dependent and information-theoretic lower bounds from a non-asymptotic viewpoint.

We propose a novel particle-based variational inference method designed to work with multimodal distributions. Our approach, referred to as Branched Stein Variational Gradient Descent (BSVGD), extends the classical Stein Variational Gradient Descent (SVGD) algorithm by incorporating a random branching mechanism that encourages the exploration of the state space. In this work, a theoretical guarantee for the convergence in distribution is presented, as well as numerical experiments to validate the suitability of our algorithm. Performance comparisons between the BSVGD and the SVGD are presented using the Wasserstein distance between samples and the corresponding computational times.

Characterizing the convergence of real-valued or vector-v alued sequences is a key theoretical problem in data science, where the sequence index typically correspon ds to the number of iterations of an iterative algorithm (such as in optimization and signal processing) o r the number of observations (as in statistics and machine learning). This characterization can be done in mostly two ways, asymptotically or non-asymptotically. In an asymptotic analysis, an asymptotic e quivalent of the sequence is identified, which readily allows comparisons with other algorithms; however, without further analysis, the behavior at any finite time cannot be controlled. This is exactly what non-as ymptotic analysis aims to achieve, by providing bounds that are valid even for a finite index, but then only pro viding bounds that cannot always be compared. While the two approaches have their own merits, in this paper, we focus on asymptotic analysis and sequences that tend to their limit at a sub-exponential r ate that is a power of the sequence index. The main goal of this paper is to show how a classical tool from signal processing, control theory, and electrical engineering ( Oppenheim et al., 1996), the z -transform method ( Jury, 1964), can be used in this context with a striking efficiency at obtaining asymptotic eq uivalents for the class of algorithms that can be seen as iterations of potentially random linear operators i n a Hilbert space. This includes gradient descent for quadratic optimization problems as well as its accelera ted and stochastic variants ( Nesterov, 2018), 1 Landweber iterations in inverse problems ( Benning and Burger, 2018), or gossip algorithms in distributed computing ( Boyd et al., 2006).

In multi-source learning with discrete labels, distributional heterogeneity across domains poses a central challenge to developing predictive models that transfer reliably to unseen domains. We study multi-source unsupervised domain adaptation, where labeled data are drawn from multiple source domains and only unlabeled data from a target domain. To address potential distribution shifts, we propose a novel Conditional Group Distributionally Robust Optimization (CG-DRO) framework that learns a classifier by minimizing the worst-case cross-entropy loss over the convex combinations of the conditional outcome distributions from the sources. To solve the resulting minimax problem, we develop an efficient Mirror Prox algorithm, where we employ a double machine learning procedure to estimate the risk function. This ensures that the errors of the machine learning estimators for the nuisance models enter only at higher-order rates, thereby preserving statistical efficiency under covariate shift. We establish fast statistical convergence rates for the estimator by constructing two surrogate minimax optimization problems that serve as theoretical bridges. A distinguishing challenge for CG-DRO is the emergence of nonstandard asymptotics: the empirical estimator may fail to converge to a standard limiting distribution due to boundary effects and system instability. To address this, we introduce a perturbation-based inference procedure that enables uniformly valid inference, including confidence interval construction and hypothesis testing.

Solutions of optimization problems, including policy optimization in reinforcement learning, typically rely upon some variant of gradient descent. There has been much recent work in the machine learning, control, and optimization communities applying the Polyak-Łojasiewicz Inequality (PLI) to such problems in order to establish an exponential rate of convergence (a.k.a. ``linear convergence'' in the local-iteration language of numerical analysis) of loss functions to their minima under the gradient flow. Often, as is the case of policy iteration for the continuous-time LQR problem, this rate vanishes for large initial conditions, resulting in a mixed globally linear / locally exponential behavior. This is in sharp contrast with the discrete-time LQR problem, where there is global exponential convergence. That gap between CT and DT behaviors motivates the search for various generalized PLI-like conditions, and this talk will address that topic. Moreover, these generalizations are key to understanding the transient and asymptotic effects of errors in the estimation of the gradient, errors which might arise from adversarial attacks, wrong evaluation by an oracle, early stopping of a simulation, inaccurate and very approximate digital twins, stochastic computations (algorithm ``reproducibility''), or learning by sampling from limited data. We describe an ``input to state stability'' (ISS) analysis of this issue. The second part discusses convergence and PLI-like properties of ``linear feedforward neural networks'' in feedback control. Much of the work described here was done in collaboration with Arthur Castello B. de Oliveira, Leilei Cui, Zhong-Ping Jiang, and Milad Siami.

We study online optimization methods for zero-sum games, a fundamental problem in adversarial learning in machine learning, economics, and many other domains. Traditional methods approximate Nash equilibria (NE) using either regret-based methods (time-average convergence) or contraction-map-based methods (last-iterate convergence). We propose a new method based on Hamiltonian dynamics in physics and prove that it can characterize the set of NE in a finite (linear) number of iterations of alternating gradient descent in the unbounded setting, modulo degeneracy, a first in online optimization. Unlike standard methods for computing NE, our proposed approach can be parallelized and works with arbitrary learning rates, both firsts in algorithmic game theory. Experimentally, we support our results by showing our approach drastically outperforms standard methods.