In this paper, we study different approaches for classifying emotions from speech using acoustic and text-based features. We propose to obtain contextualized word embeddings with BERT to represent the information contained in speech transcriptions and show that this results in better performance than using Glove embeddings. We also propose and compare different strategies to combine the audio and text modalities, evaluating them on IEMOCAP and MSP-PODCAST datasets. We find that fusing acoustic and text-based systems is beneficial on both datasets, though only subtle differences are observed across the evaluated fusion approaches. Finally, for IEMOCAP, we show the large effect that the criteria used to define the cross-validation folds have on results. In particular, the standard way of creating folds for this dataset results in a highly optimistic estimation of performance for the text-based system, suggesting that some previous works may overestimate the advantage of incorporating transcriptions.

Simulated events are key ingredients in almost all high-energy physics analyses. However, imperfections in the simulation can lead to sizeable differences between the observed data and simulated events. The effects of such mismodelling on relevant observables must be corrected either effectively via scale factors, with weights or by modifying the distributions of the observables and their correlations. We introduce a correction method that transforms one multidimensional distribution (simulation) into another one (data) using a simple architecture based on a single normalising flow with a boolean condition. We demonstrate the effectiveness of the method on a physics-inspired toy dataset with non-trivial mismodelling of several observables and their correlations.

Many real-world data sets contain a temporal component or involve transitions from state to state. For exploratory data analysis, we can represent these high-dimensional data sets in two-dimensional maps, using embeddings of the data objects under exploration and representing their temporal relationships with directed edges. Most existing dimensionality reduction techniques, such as t-SNE and UMAP, do not take into account the temporal or relational nature of the data when constructing the embeddings, resulting in temporally cluttered visualizations that obscure potentially interesting patterns. To address this problem, we propose two complementary, direction-aware loss terms in the optimization function of t-SNE that emphasize the temporal aspects of the data, guiding the optimization and the resulting embedding to reveal temporal patterns that might otherwise go unnoticed. The Directional Coherence Loss (DCL) encourages nearby arrows connecting two adjacent time series points to point in the same direction, while the Edge Length Loss (ELL) penalizes arrows - which effectively represent time gaps in the visualized embedding - based on their length. Both loss terms are differentiable and can be easily incorporated into existing dimensionality reduction techniques. By promoting local directionality of the directed edges, our procedure produces more temporally meaningful and less cluttered visualizations. We demonstrate the effectiveness of our approach on a toy dataset and two real-world datasets.

This study focuses on addressing the instability issues prevalent in contrastive learning, specifically examining the InfoNCE loss function and its derivatives. We reveal a critical observation that these loss functions exhibit a restrictive behavior, leading to a convergence phenomenon where embeddings tend to merge into a singular point. This "over-fusion" effect detrimentally affects classification accuracy in subsequent supervised-learning tasks. Through theoretical analysis, we demonstrate that embeddings, when equalized or confined to a rank-1 linear subspace, represent a local minimum for InfoNCE. In response to this challenge, our research introduces an innovative strategy that leverages the same or fewer labeled data than typically used in the fine-tuning phase. The loss we proposed, Orthonormal Anchor Regression Loss, is designed to disentangle embedding clusters, significantly enhancing the distinctiveness of each embedding while simultaneously ensuring their aggregation into dense, well-defined clusters. Our method demonstrates remarkable improvements with just a fraction of the conventional label requirements, as evidenced by our results on CIFAR10 and CIFAR100 datasets.

This paper introduces a novel K-means clustering algorithm, an advancement on the conventional Big-means methodology. The proposed method efficiently integrates parallel processing, stochastic sampling, and competitive optimization to create a scalable variant designed for big data applications. It addresses scalability and computation time challenges typically faced with traditional techniques. The algorithm adjusts sample sizes dynamically for each worker during execution, optimizing performance. Data from these sample sizes are continually analyzed, facilitating the identification of the most efficient configuration. By incorporating a competitive element among workers using different sample sizes, efficiency within the Big-means algorithm is further stimulated. In essence, the algorithm balances computational time and clustering quality by employing a stochastic, competitive sampling strategy in a parallel computing setting.

Applying iterative solvers on sparsity-constrained optimization (SCO) requires tedious mathematical deduction and careful programming/debugging that hinders these solvers' broad impact. In the paper, the library skscope is introduced to overcome such an obstacle. With skscope, users can solve the SCO by just programming the objective function. The convenience of skscope is demonstrated through two examples in the paper, where sparse linear regression and trend filtering are addressed with just four lines of code. More importantly, skscope's efficient implementation allows state-of-the-art solvers to quickly attain the sparse solution regardless of the high dimensionality of parameter space. Numerical experiments reveal the available solvers in skscope can achieve up to 80x speedup on the competing relaxation solutions obtained via the benchmarked convex solver.

Meta-learning involves training models on a variety of training tasks in a way that enables them to generalize well on new, unseen test tasks. In this work, we consider meta-learning within the framework of high-dimensional multivariate random-effects linear models and study generalized ridge-regression based predictions. The statistical intuition of using generalized ridge regression in this setting is that the covariance structure of the random regression coefficients could be leveraged to make better predictions on new tasks. Accordingly, we first characterize the precise asymptotic behavior of the predictive risk for a new test task when the data dimension grows proportionally to the number of samples per task. We next show that this predictive risk is optimal when the weight matrix in generalized ridge regression is chosen to be the inverse of the covariance matrix of random coefficients. Finally, we propose and analyze an estimator of the inverse covariance matrix of random regression coefficients based on data from the training tasks. As opposed to intractable MLE-type estimators, the proposed estimators could be computed efficiently as they could be obtained by solving (global) geodesically-convex optimization problems. Our analysis and methodology use tools from random matrix theory and Riemannian optimization. Simulation results demonstrate the improved generalization performance of the proposed method on new unseen test tasks within the considered framework.

Advances in high-throughput technologies have originated an ever-increasing availability of omics datasets. The integration of multiple heterogeneous data sources is currently an issue for biology and bioinformatics. Multiple kernel learning (MKL) has shown to be a flexible and valid approach to consider the diverse nature of multi-omics inputs, despite being an underused tool in genomic data mining.We provide novel MKL approaches based on different kernel fusion strategies.To learn from the meta-kernel of input kernels, we adaptedunsupervised integration algorithms for supervised tasks with support vector machines.We also tested deep learning architectures for kernel fusion and classification.The results show that MKL-based models can compete with more complex, state-of-the-art, supervised multi-omics integrative approaches. Multiple kernel learning offers a natural framework for predictive models in multi-omics genomic data. Our results offer a direction for bio-data mining research and further development of methods for heterogeneous data integration.

The rise in internet usage has led to the generation of massive amounts of data, resulting in the adoption of various supervised and semi-supervised machine learning algorithms, which can effectively utilize the colossal amount of data to train models. However, before deploying these models in the real world, these must be strictly evaluated on performance measures like worst-case recall and satisfy constraints such as fairness. We find that current state-of-the-art empirical techniques offer sub-optimal performance on these practical, non-decomposable performance objectives. To bridge the gap, we propose SelMix, a selective mixup-based inexpensive fine-tuning technique for pre-trained models, to optimize for the desired objective. The core idea of our framework is to determine a sampling distribution to perform a mixup of features between samples from particular classes such that it optimizes the given objective. We comprehensively evaluate our technique against the existing empirical and theoretically principled methods on standard benchmark datasets for imbalanced classification. We find that proposed SelMix fine-tuning significantly improves the performance for various practical non-decomposable objectives across benchmarks. The rise of deep networks has shown great promise by reaching near-perfect performance across computer vision tasks (He et al., 2022; Kolesnikov et al., 2020; Kirillov et al., 2023; Girdhar et al., 2023). It has led to their widespread deployment for practical applications, some of which have critical consequences (Castelvecchi, 2020). Hence, these deployed models must perform robustly across the entire data distribution and not just the majority part. These failure cases are often overlooked when considering only accuracy as our primary performance metric. Therefore, more practical metrics like Recall H-Mean (Sun et al., 2006), Worst-Case (Min) Recall (Narasimhan & Menon, 2021; Mohri et al., 2019), etc., should be used for evaluation.

This paper proposes an early detection method for cluster structural changes. Cluster structure refers to discrete structural characteristics, such as the number of clusters, when data are represented using finite mixture models, such as Gaussian mixture models. We focused on scenarios in which the cluster structure gradually changed over time. For finite mixture models, the concept of mixture complexity (MC) measures the continuous cluster size by considering the cluster proportion bias and overlap between clusters. In this paper, we propose MC fusion as an extension of MC to handle situations in which multiple mixture numbers are possible in a finite mixture model. By incorporating the fusion of multiple models, our approach accurately captured the cluster structure during transitional periods of gradual change. Moreover, we introduce a method for detecting changes in the cluster structure by examining the transition of MC fusion. We demonstrate the effectiveness of our method through empirical analysis using both artificial and real-world datasets.