Data & Analytics

**arXiv ID:** 1902.01838 **Authors:** Amritanshu Agrawal, Wei Fu, Di Chen, Xipeng Shen, Tim Menzies **Published:** 2019-02-05T18:16:56Z **Abstract:** Machine learning techniques applied to software engineering tasks can be improved by hyperparameter optimization, i.e., automatic tools that find good settings for a learner's control parameters. We show that such hyperparameter optimization can be unnecessarily slow, particularly when the optimizers waste time exploring "redundant tunings"', i....

**arXiv ID:** 2512.04475 **Authors:** Timo Stoll, Chendi Qian, Ben Finkelshtein, Ali Parviz, Darius Weber, Fabrizio Frasca, Hadar Shavit, Antoine Siraudin, Arman Mielke, Marie Anastacio, Erik Müller, Maya Bechler-Speicher, Michael Bronstein, Mikhail Galkin, Holger Hoos, Mathias Niepert, Bryan Perozzi, Jan Tönshoff, Christopher Morris **Published:** 2025-12-04T05:30:31Z **Abstract:** Machine learning on graphs has made substantial progress across domains such as molecular property prediction a...

**arXiv ID:** 2212.04458 **Authors:** Louis Kirsch, James Harrison, Jascha Sohl-Dickstein, Luke Metz **Published:** 2022-12-08T18:30:22Z **Abstract:** Modern machine learning requires system designers to specify aspects of the learning pipeline, such as losses, architectures, and optimizers. Meta-learning, or learning-to-learn, instead aims to learn those aspects, and promises to unlock greater capabilities with less manual effort. One particularly ambitious goal of meta-learning is to train ...

**arXiv ID:** 2303.01584 **Authors:** Noah Barrett, Zahra Sadeghi, Stan Matwin **Published:** 2023-03-02T21:16:53Z **Abstract:** Self-supervised Learning (SSL) is a machine learning algorithm for pretraining Deep Neural Networks (DNNs) without requiring manually labeled data. The central idea of this learning technique is based on an auxiliary stage aka pretext task in which labeled data are created automatically through data augmentation and exploited for pretraining the DNN. However, the ef...

Energy-based dynamical systems framework for neurocomputation, learning, and optimization. Unifies Hopfield networks, Boltzmann machines, modern EBMs, and equilibrium propagation under a single energy landscape formulation. Covers gradient flow dynamics, attractor analysis, contrastive learning, and biologically-plausible learning rules. Activation: energy-based models, EBMs, neural dynamics, Hopfield networks, energy landscape, attractor dynamics, gradient flow, equilibrium propagation, cont...

Recent advances at the intersection of control theory, neuroscience, and machine learning have revealed novel mechanisms by which dynamical systems perform computation. These advances encompass a wide. Activation: energy-based models, dynamical systems, ODE complexity

Theoretical analysis of effective target shift in online learning and methods to correct for it. Explains why online learning struggles under distributional shift and how to characterize the relationship between online and offline learning. Activation triggers: online learning, target shift, distributional shift, online vs offline learning, sequential learning theory

Effective rank methodology for predicting quantum data encoding performance. Uses feature map effective rank as a threshold criterion to accelerate the search for high-performing QML encodings. Activation: effective rank encoding, feature map rank QML, encoding performance prediction, quantum encoding predictor, QML encoding ranking.

Distributional matrix completion methodology using kernel mean embeddings and Tucker rank for probability-distribution-valued matrices. Represents each matrix entry as a probability distribution via RKHS embeddings, introduces functional unfolding operators to bridge infinite-dimensional embeddings with finite-dimensional tensor structure. Applicable to statistical learning with distributional data, quantum state tomography, financial risk modeling.

Methodology for auditing and distilling deep reinforcement learning hedging policies into interpretable symbolic formulas. Includes framework for analyzing delta corrections relative to Black-Scholes, symbolic regression distillation, and regime fragility stress-testing. Use when analyzing neural hedging strategies, quantitative risk management, options hedging with RL, or making black-box financial AI auditable.

Convolutional Neural Network and Adversarial Autoencoder in EEG images classification... Activation: adversarial, 脑电图, 对抗, eeg, 脑

... Activation: EEG, brain signal, electroencephalography, brain network, graph, connectivity

Convergent representations of linguistic constructions in human and artificial neural systems. Analyzes alignment between biological brain activity (EEG) and artificial neural language models (RNNs, Transformers) in processing Argument Structure Constructions. Activation: linguistic constructions, ASC, brain-language alignment, EEG language, construction grammar, convergent representations.

**arXiv ID:** 2406.02597 **Authors:** Karn Tiwari, N M Anoop Krishnan, A P Prathosh **Published:** 2024-06-01T14:32:19Z **Abstract:** Neural operators extend data-driven models to map between infinite-dimensional functional spaces. While these operators perform effectively in either the time or frequency domain, their performance may be limited when applied to non-stationary spatial or temporal signals whose frequency characteristics change with time. Here, we introduce Complex Neural Operato...

**arXiv ID:** 1804.02341 **Authors:** Edward Choi, Angeliki Lazaridou, Nando de Freitas **Published:** 2018-04-06T16:12:51Z **Abstract:** One of the distinguishing aspects of human language is its compositionality, which allows us to describe complex environments with limited vocabulary. Previously, it has been shown that neural network agents can learn to communicate in a highly structured, possibly compositional language based on disentangled input (e.g. hand- engineered features). Humans, ...

One-shot federated learning (OSFL) addresses the communication overhead of federated learning by limiting training to a single round, but doing so without sacrificing model quality is non-trivial, par. Based on arXiv:2607.07565.

Cold-atom (neutral-atom) reservoir computing methodology for efficient machine learning tasks. Uses Rydberg atom arrays as physical reservoirs, encoding input data into Hamiltonian parameters and reading out via quantum measurements. Use when implementing reservoir computing on quantum hardware, exploring neutral-atom ML platforms, or building energy-efficient quantum-inspired classifiers. Triggers: cold atom reservoir, neutral atom computing, Rydberg reservoir, quantum reservoir machine, ato...

**arXiv ID:** 2409.14429 **Authors:** Sven Kruschel, Nico Hambauer, Sven Weinzierl, Sandra Zilker, Mathias Kraus, Patrick Zschech **Published:** 2024-09-22T12:58:52Z **Abstract:** Machine learning is permeating every conceivable domain to promote data-driven decision support. The focus is often on advanced black-box models due to their assumed performance advantages, whereas interpretable models are often associated with inferior predictive qualities. More recently, however, a new generation ...

Lightweight self-supervised representation learning for fMRI using positive-only data pairs, achieving strong cross-task generalization without large-scale pretraining

Two-stage interpretation of attention as in-context empirical Bayes inference via particle dynamics with posterior mean recovery guarantees

**arXiv ID:** 1708.00339 **Authors:** Ritambhara Singh, Jack Lanchantin, Arshdeep Sekhon, Yanjun Qi **Published:** 2017-08-01T14:06:12Z **Abstract:** The past decade has seen a revolution in genomic technologies that enable a flood of genome-wide profiling of chromatin marks. Recent literature tried to understand gene regulation by predicting gene expression from large-scale chromatin measurements. Two fundamental challenges exist for such learning tasks: (1) genome-wide chromatin signals are...

**arXiv ID:** 2306.00040 **Authors:** Ana Nikolikj, Gjorgjina Cenikj, Gordana Ispirova, Diederick Vermetten, Ryan Dieter Lang, Andries Petrus Engelbrecht, Carola Doerr, Peter Korošec, Tome Eftimov **Published:** 2023-05-31T12:50:44Z **Abstract:** A key component of automated algorithm selection and configuration, which in most cases are performed using supervised machine learning (ML) methods is a good-performing predictive model. The predictive model uses the feature representation of a set ...

**arXiv ID:** 2609.10012v1 **Authors:** Fatemeh Mahmoudi **URL:** http://arxiv.org/abs/2609.10012v1 **Utility Score:** 1.00

**arXiv ID:** 2608.25759v1 **Authors:** Hilda Adwubi Osei, Catherine Tenewaa Osei, Desdemona Yaa Asobayire **URL:** http://arxiv.org/abs/2608.25759v1 **Utility Score:** 1.00