The paper introduces a framework that learns the kernel used in kernel methods through alignment, leveraging the Collaborative Learning and Inference (CLaI) approach. It demonstrates that CLaI can be interpreted as a kernel alignment process and that its inference stage is equivalent to kernel Bayes classification with Parzen-window density estimation. By replacing cosine similarity with a learned Mahalanobis distance, the authors extend CLaI to multiclass classification, achieving higher accuracy, faster convergence, and lower calibration error on datasets such as CIFAR-10, PathMNIST, and SleepEDF, while also showing connections to Gaussian processes and competitive calibration in sepsis prediction.
By Hollan Haule, Alfredo Gonzalez-Sulser, Javier Escudero
arXiv:2509. 12760v5 Announce Type: replace Abstract: We introduce the Similarity-Distance-Magnitude (SDM) activation function, a more robust and interpretable formulation of the standard softmax activation function, adding Similarity (i.
By Allen Schmaltz
The paper investigates why prescribed‑parameter convolutional kernels, such as those used in ROCKET and its variants, perform well on time‑series classification. It analyzes the proportion of positive values (PPV) pooling operator, proving exact invariance to circular translations and quantifying deviations under finite boundaries. Experiments on the UCR collection show that increasing kernel count yields diminishing returns, while longer kernels raise effective dimensionality and distance correlation, with optimal test performance at intermediate lengths.
By Jorge Marco Blanco, Rub\'en Cuevas Rum\'in
arXiv:2510. 25599v2 Announce Type: replace Abstract: Regression tasks, notably in safety-critical domains, require reliable uncertainty quantification, yet the literature remains largely classification-focused.
By Christopher B\"ulte, Yusuf Sale, Gitta Kutyniok, Eyke H\"ullermeier
arXiv:2610.22564v1 Announce Type: new
Abstract: Data-driven approaches often face ambiguity in datasets due to sensor or target noise, insufficient information, and noise in the input data. While bes...
By Nina Maria Gottschling, Jakob Gawlikowski, Michele Caprio
arXiv:2606. 27731v1 Announce Type: cross Abstract: Despite their strong general capabilities, large language models (LLMs) often remain unreliable when outputs must be numerically precise.
By Zhuo Zuo, Li Yue, Wenhao Zheng, Chenpeng Wang, Xianggen Liu
arXiv:2601. 11670v3 Announce Type: replace-cross Abstract: Pseudo-label selection in semi-supervised learning is commonly driven by maximum-confidence thresholds, yet confidence alone can be unreliable under model overconfidence and class imbalance.
By Jinshi Liu, Lei He, Pan Liu
arXiv:2509. 05130v2 Announce Type: replace Abstract: In classification problems, models are trained to predict a class label based on the input data features.
By Davide Pirovano, Federico Milanesio, Michele Caselle, Piero Fariselli, Matteo Osella
arXiv:2610.00851v1 Announce Type: new
Abstract: Open Set Recognition (OSR) aims to enable models to accurately classify known classes while rejecting samples from unseen classes. A key challenge in O...
By Thiru Thillai Nadarasar Bahavan, Yu Xia, Sachith Seneviratne, Saman Halgamuge
arXiv:2606. 06335v1 Announce Type: new Abstract: Performance estimation under distribution shift aims to predict how a model behaves on an unlabeled test set whose distribution differs from the training data, a scenario that requires reliable indicators that can faithfully reflect model behavior without ground-truth labels.
By Shuxuan Li, Zhilin Zhao, Quyu Kong, Wei-Shi Zheng
arXiv:2608. 20326v1 Announce Type: cross Abstract: Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions.
By Parampreet Singh, Anushka Singh, Sumit Kumar, Vipul Arora
Deep neural networks are often overconfident, assigning high confidence even to incorrect predictions. Consequently, users lack a reliable signal for deciding when a prediction can be trusted.