Machine learning (ML) and artificial intelligence (AI) can provide powerful tools for the scientific community, as demonstrated by the recent Nobel Prize in Chemistry. Reversely, insights from traditional physics theories also contribute to a deeper understanding of the mechanism of learning. Ref. [1] contains a broad overview of the successful cross-fertilisation between ML and the physical sciences, covering a number of domains. One way to mitigate against possible scepticism with regard to using ML as a "black box" is by unveiling the dynamics of training (or learning) and explaining how the relevant information is engraved in the model during the training stage. To further develop this programme, we study here the dynamics of first-order stochastic gradient descent as applied to weight matrices, reporting and expanding on the work presented in Ref. [2]. When training ML models, weight matrices are commonly updated by one of the variants of the stochastic gradient descent algorithm. The dynamics can then be decomposed into a drift and a fluctuating term, and such a system can be described by a discrete Langevin equation. The dynamics of stochastic matrix updates is richer than the dynamics for vector or scalar quantities, as captured by Dyson Brownian motion and random matrix theory (RMT), with the appearance of universal features for the eigenvalues [3-9]. Earlier descriptions of the statistical properties of weight matrices in terms of RMT can be found in e.g.

Recent findings by Cohen et al., 2021, demonstrate that when training neural networks with full-batch gradient descent at a step size of $\eta$, the sharpness--defined as the largest eigenvalue of the full batch Hessian--consistently stabilizes at $2/\eta$. These results have significant implications for convergence and generalization. Unfortunately, this was observed not to be the case for mini-batch stochastic gradient descent (SGD), thus limiting the broader applicability of these findings. We show that SGD trains in a different regime we call Edge of Stochastic Stability. In this regime, what hovers at $2/\eta$ is, instead, the average over the batches of the largest eigenvalue of the Hessian of the mini batch (MiniBS) loss--which is always bigger than the sharpness. This implies that the sharpness is generally lower when training with smaller batches or bigger learning rate, providing a basis for the observed implicit regularization effect of SGD towards flatter minima and a number of well established empirical phenomena. Additionally, we quantify the gap between the MiniBS and the sharpness, further characterizing this distinct training regime.

Single-cell RNA sequencing allows the quantitation of gene expression at the individual cell level, enabling the study of cellular heterogeneity and gene expression dynamics. Dimensionality reduction is a common preprocessing step to simplify the visualization, clustering, and phenotypic characterization of samples. This step, often performed using principal component analysis or closely related methods, is challenging because of the size and complexity of the data. In this work, we present a generalized matrix factorization model assuming a general exponential dispersion family distribution and we show that many of the proposed approaches in the single-cell dimensionality reduction literature can be seen as special cases of this model. Furthermore, we propose a scalable adaptive stochastic gradient descent algorithm that allows us to estimate the model efficiently, enabling the analysis of millions of cells. Our contribution extends to introducing a novel warm start initialization method, designed to accelerate algorithm convergence and increase the precision of final estimates. Moreover, we discuss strategies for dealing with missing values and model selection. We benchmark the proposed algorithm through extensive numerical experiments against state-of-the-art methods and showcase its use in real-world biological applications. The proposed method systematically outperforms existing methods of both generalized and non-negative matrix factorization, demonstrating faster execution times while maintaining, or even enhancing, matrix reconstruction fidelity and accuracy in biological signal extraction. Finally, all the methods discussed here are implemented in an efficient open-source R package, sgdGMF, available at github/CristianCastiglione/sgdGMF

Regularization is a widely recognized technique in mathematical optimization. It can be used to smooth out objective functions, refine the feasible solution set, or prevent overfitting in machine learning models. Due to its simplicity and robustness, the gradient descent (GD) method is one of the primary methods used for numerical optimization of differentiable objective functions. However, GD is not well-suited for solving $\ell^1$ regularized optimization problems since these problems are non-differentiable at zero, causing iteration updates to oscillate or fail to converge. Instead, a more effective version of GD, called the proximal gradient descent employs a technique known as soft-thresholding to shrink the iteration updates toward zero, thus enabling sparsity in the solution. Motivated by the widespread applications of proximal GD in sparse and low-rank recovery across various engineering disciplines, we provide an overview of the GD and proximal GD methods for solving regularized optimization problems. Furthermore, this paper proposes a novel algorithm for the proximal GD method that incorporates a variable step size. Unlike conventional proximal GD, which uses a fixed step size based on the global Lipschitz constant, our method estimates the Lipschitz constant locally at each iteration and uses its reciprocal as the step size. This eliminates the need for a global Lipschitz constant, which can be impractical to compute. Numerical experiments we performed on synthetic and real-data sets show notable performance improvement of the proposed method compared to the conventional proximal GD with constant step size, both in terms of number of iterations and in time requirements.

Federated Learning (FL) has received much attention in recent years. However, although clients are not required to share their data in FL, the global model itself can implicitly remember clients' local data. Therefore, it's necessary to effectively remove the target client's data from the FL global model to ease the risk of privacy leakage and implement ``the right to be forgotten". Federated Unlearning (FU) has been considered a promising way to remove data without full retraining. But the model utility easily suffers significant reduction during unlearning due to the gradient conflicts. Furthermore, when conducting the post-training to recover the model utility, the model is prone to move back and revert what has already been unlearned. To address these issues, we propose Federated Unlearning with Orthogonal Steepest Descent (FedOSD). We first design an unlearning Cross-Entropy loss to overcome the convergence issue of the gradient ascent. A steepest descent direction for unlearning is then calculated in the condition of being non-conflicting with other clients' gradients and closest to the target client's gradient. This benefits to efficiently unlearn and mitigate the model utility reduction. After unlearning, we recover the model utility by maintaining the achievement of unlearning. Finally, extensive experiments in several FL scenarios verify that FedOSD outperforms the SOTA FU algorithms in terms of unlearning and model utility.

This paper studies the problem of stochastic bilevel optimization where the upper-level function is nonconvex with potentially unbounded smoothness and the lower-level function is strongly convex. This problem is motivated by meta-learning applied to sequential data, such as text classification using recurrent neural networks, where the smoothness constant of the upper-level loss function scales linearly with the gradient norm and can be potentially unbounded. Existing algorithm crucially relies on the nested loop design, which requires significant tuning efforts and is not practical. In this paper, we address this issue by proposing a Single Loop bIlevel oPtimizer (SLIP). The proposed algorithm first updates the lower-level variable by a few steps of stochastic gradient descent, and then simultaneously updates the upper-level variable by normalized stochastic gradient descent with momentum and the lower-level variable by stochastic gradient descent. Under standard assumptions, we show that our algorithm finds an $\epsilon$-stationary point within $\widetilde{O}(1/\epsilon^4)$\footnote{Here $\widetilde{O}(\cdot)$ compresses logarithmic factors of $1/\epsilon$ and $1/\delta$, where $\delta\in(0,1)$ denotes the failure probability.} oracle calls of stochastic gradient or Hessian-vector product, both in expectation and with high probability. This complexity result is nearly optimal up to logarithmic factors without mean-square smoothness of the stochastic gradient oracle. Our proof relies on (i) a refined characterization and control of the lower-level variable and (ii) establishing a novel connection between bilevel optimization and stochastic optimization under distributional drift. Our experiments on various tasks show that our algorithm significantly outperforms strong baselines in bilevel optimization.

Stochastic Gradient Descent (SGD) with gradient clipping is a powerful technique for enabling differentially private optimization. Although prior works extensively investigated clipping with a constant threshold, private training remains highly sensitive to threshold selection, which can be expensive or even infeasible to tune. This sensitivity motivates the development of adaptive approaches, such as quantile clipping, which have demonstrated empirical success but lack a solid theoretical understanding. This paper provides the first comprehensive convergence analysis of SGD with quantile clipping (QC-SGD). We demonstrate that QC-SGD suffers from a bias problem similar to constant-threshold clipped SGD but show how this can be mitigated through a carefully designed quantile and step size schedule. Our analysis reveals crucial relationships between quantile selection, step size, and convergence behavior, providing practical guidelines for parameter selection. We extend these results to differentially private optimization, establishing the first theoretical guarantees for DP-QC-SGD. Our findings provide theoretical foundations for widely used adaptive clipping heuristic and highlight open avenues for future research.

Recently, the study of heavy-tailed noises in first-order nonconvex stochastic optimization has gotten a lot of attention since it was recognized as a more realistic condition as suggested by many empirical observations. Specifically, the stochastic noise (the difference between the stochastic and true gradient) is considered only to have a finite $\mathfrak{p}$-th moment where $\mathfrak{p}\in\left(1,2\right]$ instead of assuming it always satisfies the classical finite variance assumption. To deal with this more challenging setting, people have proposed different algorithms and proved them to converge at an optimal $\mathcal{O}(T^{\frac{1-\mathfrak{p}}{3\mathfrak{p}-2}})$ rate for smooth objectives after $T$ iterations. Notably, all these new-designed algorithms are based on the same technique - gradient clipping. Naturally, one may want to know whether the clipping method is a necessary ingredient and the only way to guarantee convergence under heavy-tailed noises. In this work, by revisiting the existing Batched Normalized Stochastic Gradient Descent with Momentum (Batched NSGDM) algorithm, we provide the first convergence result under heavy-tailed noises but without gradient clipping. Concretely, we prove that Batched NSGDM can achieve the optimal $\mathcal{O}(T^{\frac{1-\mathfrak{p}}{3\mathfrak{p}-2}})$ rate even under the relaxed smooth condition. More interestingly, we also establish the first $\mathcal{O}(T^{\frac{1-\mathfrak{p}}{2\mathfrak{p}}})$ convergence rate in the case where the tail index $\mathfrak{p}$ is unknown in advance, which is arguably the common scenario in practice.

This study presents an autonomous experimental machine learning protocol for high-frequency trading (HFT) stock price forecasting that involves a dual competitive feature importance mechanism and clustering via shallow neural network topology for fast training. By incorporating the k-means algorithm into the radial basis function neural network (RBFNN), the proposed method addresses the challenges of manual clustering and the reliance on potentially uninformative features. More specifically, our approach involves a dual competitive mechanism for feature importance, combining the mean-decrease impurity (MDI) method and a gradient descent (GD) based feature importance mechanism. This approach, tested on HFT Level 1 order book data for 20 S&P 500 stocks, enhances the forecasting ability of the RBFNN regressor. Our findings suggest that an autonomous approach to feature selection and clustering is crucial, as each stock requires a different input feature space. Overall, by automating the feature selection and clustering processes, we remove the need for manual topological grid search and provide a more efficient way to predict LOB's mid-price.

Due to the non-smoothness of optimization problems in Machine Learning, generalized smoothness assumptions have been gaining a lot of attention in recent years. One of the most popular assumptions of this type is $(L_0,L_1)$-smoothness (Zhang et al., 2020). In this paper, we focus on the class of (strongly) convex $(L_0,L_1)$-smooth functions and derive new convergence guarantees for several existing methods. In particular, we derive improved convergence rates for Gradient Descent with (Smoothed) Gradient Clipping and for Gradient Descent with Polyak Stepsizes. In contrast to the existing results, our rates do not rely on the standard smoothness assumption and do not suffer from the exponential dependency from the initial distance to the solution. We also extend these results to the stochastic case under the over-parameterization assumption, propose a new accelerated method for convex $(L_0,L_1)$-smooth optimization, and derive new convergence rates for Adaptive Gradient Descent (Malitsky and Mishchenko, 2020).