This article examines the implicit regularization effect of Stochastic Gradient Descent (SGD). We consider the case of SGD without replacement, the variant typically used to optimize large-scale neural networks. We analyze this algorithm in a more realistic regime than typically considered in theoretical works on SGD, as, e.g., we allow the product of the learning rate and Hessian to be $O(1)$. Our core theoretical result is that optimizing with SGD without replacement is locally equivalent to making an additional step on a novel regularizer. This implies that the trajectory of SGD without replacement diverges from both noise-injected GD and SGD with replacement (in which batches are sampled i.i.d.). Indeed, the two SGDs travel flat regions of the loss landscape in distinct directions and at different speeds. In expectation, SGD without replacement may escape saddles significantly faster and present a smaller variance. Moreover, we find that SGD implicitly regularizes the trace of the noise covariance in the eigendirections of small and negative Hessian eigenvalues. This coincides with penalizing a weighted trace of the Fisher Matrix and the Hessian on several vision tasks, thus encouraging sparsity in the spectrum of the Hessian of the loss in line with empirical observations from prior work. We also propose an explanation for why SGD does not train at the edge of stability (as opposed to GD).

At the crossway of machine learning and data analysis, anomaly detection aims at identifying observations that exhibit abnormal behaviour. Be it measurement errors, disease development, severe weather, production quality default(s) (items) or failed equipment, financial frauds or crisis events, their on-time identification and isolation constitute an important task in almost any area of industry and science. While a substantial body of literature is devoted to detection of anomalies, little attention is payed to their explanation. This is the case mostly due to intrinsically non-supervised nature of the task and non-robustness of the exploratory methods like principal component analysis (PCA). We introduce a new statistical tool dedicated for exploratory analysis of abnormal observations using data depth as a score. Anomaly component analysis (shortly ACA) is a method that searches a low-dimensional data representation that best visualises and explains anomalies. This low-dimensional representation not only allows to distinguish groups of anomalies better than the methods of the state of the art, but as well provides a -- linear in variables and thus easily interpretable -- explanation for anomalies. In a comparative simulation and real-data study, ACA also proves advantageous for anomaly analysis with respect to methods present in the literature.

Continuously-observed event occurrences, often exhibit self- and mutually-exciting effects, which can be well modeled using temporal point processes. Beyond that, these event dynamics may also change over time, with certain periodic trends. We propose a novel variational auto-encoder to capture such a mixture of temporal dynamics. More specifically, the whole time interval of the input sequence is partitioned into a set of sub-intervals. The event dynamics are assumed to be stationary within each sub-interval, but could be changing across those sub-intervals. In particular, we use a sequential latent variable model to learn a dependency graph between the observed dimensions, for each sub-interval. The model predicts the future event times, by using the learned dependency graph to remove the noncontributing influences of past events. By doing so, the proposed model demonstrates its higher accuracy in predicting inter-event times and event types for several real-world event sequences, compared with existing state of the art neural point processes.

The Tree of Life is a massive graphical structure which represents the evolutionary process from single cell organisms into the immense biodiversity of living species in present time. Estimating the Tree of Life would not only represent the greatest accomplishment in evolutionary biology and systematics, but it would also allow us to fully understand the development and evolution of important biological traits in nature, in particular, those related to resilience to extinction when exposed to environmental threats such as climate change. Therefore, the development of statistical and machine-learning theory to reconstruct the Tree of Life, especially those scalable to big data, are paramount in evolutionary biology, systematics, and conservation efforts against mass extinctions. Graphical structures that represent evolutionary processes are denoted phylogenetic trees. A phylogenetic tree is a binary tree whose internal nodes represent ancestral species that over time differentiate into two separate species giving rise to its two children nodes (see Figure 1 left). The evolutionary process is then depicted by this bifurcating tree from the root (the origin of life) to the external nodes of the tree (also denoted leaves) which represent the living organisms today.

This article presents a new polynomial parameterized sigmoid called SIGTRON, which is an extended asymmetric sigmoid with Perceptron, and its companion convex model called SIGTRON-imbalanced classification (SIC) model that employs a virtual SIGTRON-induced convex loss function. In contrast to the conventional π-weighted costsensitive learning model, the SIC model does not have an external π-weight on the loss function but has internal parameters in the virtual SIGTRON-induced loss function. As a consequence, when the given training dataset is close to the well-balanced condition, we show that the proposed SIC model is more adaptive to variations of the dataset, such as the inconsistency of the scale-class-imbalance ratio between the training and test datasets. This adaptation is achieved by creating a skewed hyperplane equation. Additionally, we present a quasi-Newton optimization(L-BFGS) framework for the virtual convex loss by developing an interval-based bisection line search. Empirically, we have observed that the proposed approach outperforms π-weighted convex focal loss and balanced classifier LIBLINEAR(logistic regression, SVM, and L2SVM) in terms of test classification accuracy with 51 two-class and 67 multi-class datasets. In binary classification problems, where the scale-class-imbalance ratio of the training dataset is not significant but the inconsistency exists, a group of SIC models with the best test accuracy for each dataset (TOP1) outperforms LIBSVM(C-SVC with RBF kernel), a well-known kernel-based classifier. The main hindrance of the process is that the dataset is imbalanced [1], [2], [3] and inconsistent [4]. It is worth noting that we can improve the scale imbalance through various normalization methods [6], [7]. In our experiments, we use the well-organized datasets in [8].

Survival Analysis (SA) is about modeling the time for an event of interest to occur, which has important applications in many fields, including medicine, defense, finance, and aerospace. Recent work has demonstrated the benefits of using Neural Networks (NNs) to capture complicated relationships in SA. However, the datasets used to train these models are often subject to uncertainty (e.g., noisy measurements, human error), which we show can substantially degrade the performance of existing techniques. To address this issue, this work leverages recent advances in NN verification to provide new algorithms for generating fully parametric survival models that are robust to such uncertainties. In particular, we introduce a robust loss function for training the models and use CROWN-IBP regularization to address the computational challenges with solving the resulting Min-Max problem. To evaluate the proposed approach, we apply relevant perturbations to publicly available datasets in the SurvSet repository and compare survival models against several baselines. We empirically show that Survival Analysis with Adversarial Regularization (SAWAR) method on average ranks best for dataset perturbations of varying magnitudes on metrics such as Negative Log Likelihood (NegLL), Integrated Brier Score (IBS), and Concordance Index (CI), concluding that adversarial regularization enhances performance in SA. Code: https://github.com/mlpotter/SAWAR

We introduce a suite of new particle-based algorithms for sampling in constrained domains which are entirely learning rate free. Our approach leverages coin betting ideas from convex optimisation, and the viewpoint of constrained sampling as a mirrored optimisation problem on the space of probability measures. Based on this viewpoint, we also introduce a unifying framework for several existing constrained sampling algorithms, including mirrored Langevin dynamics and mirrored Stein variational gradient descent. We demonstrate the performance of our algorithms on a range of numerical examples, including sampling from targets on the simplex, sampling with fairness constraints, and constrained sampling problems in post-selection inference. Our results indicate that our algorithms achieve competitive performance with existing constrained sampling methods, without the need to tune any hyperparameters.

We introduce a new approach for generating sequences of implied volatility (IV) surfaces across multiple assets that is faithful to historical prices. We do so using a combination of functional data analysis and neural stochastic differential equations (SDEs) combined with a probability integral transform penalty to reduce model misspecification. We demonstrate that learning the joint dynamics of IV surfaces and prices produces market scenarios that are consistent with historical features and lie within the sub-manifold of surfaces that are essentially free of static arbitrage. Finally, we demonstrate that delta hedging using the simulated surfaces generates profit and loss (P&L) distributions that are consistent with realised P&Ls.

This study leverages graph neural networks to integrate MELC data with Radiomic-extracted features for melanoma classification, focusing on cell-wise analysis. It assesses the effectiveness of gene expression profiles and Radiomic features, revealing that Radiomic features, particularly when combined with UMAP for dimensionality reduction, significantly enhance classification performance. Notably, using Radiomics contributes to increased diagnostic accuracy and computational efficiency, as it allows for the extraction of critical data from fewer stains, thereby reducing operational costs. This methodology marks an advancement in computational dermatology for melanoma cell classification, setting the stage for future research and potential developments.

Industry practitioners care about small improvements in malware detection accuracy because their models are deployed to hundreds of millions of machines, meaning a 0.1\% change can cause an overwhelming number of false positives. However, academic research is often restrained to public datasets on the order of ten thousand samples and is too small to detect improvements that may be relevant to industry. Working within these constraints, we devise an approach to generate a benchmark of configurable difficulty from a pool of available samples. This is done by leveraging malware family information from tools like AVClass to construct training/test splits that have different generalization rates, as measured by a secondary model. Our experiments will demonstrate that using a less accurate secondary model with disparate features is effective at producing benchmarks for a more sophisticated target model that is under evaluation. We also ablate against alternative designs to show the need for our approach.