Data & Analytics

**arXiv ID:** 2105.10585 **Authors:** Philip M. Long **Published:** 2021-05-21T21:50:18Z **Abstract:** The Neural Tangent Kernel (NTK) is the wide-network limit of a kernel defined using neural networks at initialization, whose embedding is the gradient of the output of the network with respect to its parameters. We study the "after kernel", which is defined using the same embedding, except after training, for neural networks with standard architectures, on binary classification problems extr...

Post-Recurrent Module (PRM) for explainable RNN-based P300 classification in BCIs — combines performance improvement with global/local explainability techniques for transparent EEG-based neural decoding. Activation triggers: PRM, P300 BCI, explainable RNN, EEG explainability, post-recurrent module, P300 classification, transparent BCI.

PRISM (Probabilistic Recurrent Intention Switching Model) methodology for multi-intention inverse reinforcement learning. Uses lightweight recurrent networks for intention switching with closed-form EM solution. Activation: 多意图 IRL, intention switching, PRISM, 目标切换, recurrent intention, EM algorithm.

Bayesian prior elicitation methodology for single-subject functional connectivity network inference from resting-state fMRI. Introduces novel Bayesian priors on correlation matrices with a dedicated elicitation framework that translates expert beliefs about expected correlation levels and variability into interpretable hyperparameters. Provides distributional (not point) estimates of connectivity weights with uncertainty quantification and credible sets. Use when performing Bayesian functiona...

Cohomological structure analysis methodology for prime numbers — iterative maps predicting prime growth, cohomological equation solutions, and connections between statistical mechanics, quantum mechanics, and number theory. The logarithmic integral function emerges as the solution to the cohomological equation governing prime distribution.

Cohomological structure analysis of prime numbers using iterative maps, linking prime irregularities to physical systems including statistical mechanics and quantum mechanics.

**arXiv ID:** 2403.09680 **Authors:** Jordan Morris **Published:** 2024-02-07T15:30:23Z **Abstract:** This paper proposes a machine learning pre-sort stage to traditional supervised learning using Tsetlin Machines. Initially, K data-points are identified from the dataset using an expedited genetic algorithm to solve the maximum dispersion problem. These are then used as the initial placement to run the K-Medoid clustering algorithm. Finally, an expedited genetic algorithm is used to align K i...

Preisach Attention Layer (PAL) — a novel sequence modeling architecture that replaces softmax attention with the classical Preisach hysteresis operator from mathematical physics. Uses binary relay operators with learned thresholds and a stack of local extrema as internal state. Achieves Turing-completeness at O(1) depth via two-stack PDA simulation. Activation: attention, hysteresis, sequence modeling, episodic memory, transformer alternative, rate-independent computation

Multi-signal model of adult language learning using transformer brain alignment. Prediction shapes group-level neural architecture, feedback explains individual differences. fMRI-based with 102 subjects over 7 days. Activation: language learning, predictive coding, feedback signals, brain-model alignment, individual differences, transformer language models, artificial language learning, fMRI language representation.

Krylov Mean-Field Chaos theory for random recurrent networks — demonstrating that deterministic chaos has latent predictability through Krylov state space decomposition. Extends Hamiltonian chaos concepts to classical dissipative systems.

**arXiv ID:** 2212.00671 **Authors:** Anupam Biswas **Published:** 2022-12-01T17:21:44Z **Abstract:** The performance of individual evolutionary optimization algorithms is mostly measured in terms of statistics such as mean, median and standard deviation etc., computed over the best solutions obtained with few trails of the algorithm. To compare the performance of two algorithms, the values of these statistics are compared instead of comparing the solutions directly. This kind of comparison l...

A型钾电流介导的神经元增益控制机制。研究IA作为减法抑制与除法抑制之间的开关，通过动力学系统分析理解神经元如何自调节抑制效果。适用于计算神经科学、神经元建模、增益控制研究。触发词：A型钾电流、增益控制、抑制模式、IA电流、神经元增益、divisive inhibition、subtractive inhibition、gain control、potassium current。

Methodology for forecasting conscious choices using the Positive Experience Principle (PEP) derived from the Universal Consciousness Code theory.

**arXiv ID:** 2502.14546 **Authors:** Maya Bechler-Speicher, Ben Finkelshtein, Fabrizio Frasca, Luis Müller, Jan Tönshoff, Antoine Siraudin, Viktor Zaverkin, Michael M. Bronstein, Mathias Niepert, Bryan Perozzi, Mikhail Galkin, Christopher Morris **Published:** 2025-02-20T13:21:47Z **Abstract:** While machine learning on graphs has demonstrated promise in drug design and molecular property prediction, significant benchmarking challenges hinder its further progress and relevance. Current bench...

Poisson Matrix-Normal Latent Variable (PMNLV) model for partitioning neural co-variability in population recordings. Extends single-neuron overdispersion to populations with Kronecker-factored covariance for structured gain-modulation analysis. Use when analyzing neural population co-variability, overdispersion in spiking data, Neuropixel recordings, structured gain covariance, or trial-to-trial variability beyond scalar Fano factor summaries. Activation: PMNLV, neural co-variability, overdis...

Physics-informed Neural Networks (PINNs) for biomedical modeling and simulation. Use when working on physics-guided neural network approaches for hemodynamics, cardiovascular modeling, blood flow prediction, or inverse medical physics problems. Combines physical principles with neural networks for personalized medical predictions with minimal data requirements.

Physics-guided neural network design and training methods. Embed physical laws, constraints, and symmetries into neural network architecture for improved modeling of physical systems (quantum mechanics, statistical physics, fluid dynamics, materials science). Activation: physics guided neural network, 物理学指导神经网络, physics-informed neural network, PINN, physics-constrained learning, quantum neural network, physics-aware training.

PEM-UDE methodology for discovering governing equations from chaotic neural systems. Combines prediction-error method with universal differential equations to extract interpretable mathematical expressions from chaotic dynamical systems, applied to neural population dynamics. Activation: pem-ude, governing equations neural, chaotic system discovery, universal differential equations neural, symbolic regression neural, neural population dynamics discovery.

Pathwise approach to metastability for Galves-Löcherbach (GL) stochastic spiking neural network models. Reviews metastability theory from chemistry to probability theory, provides general definition encompassing GL model variants, surveys established metastability results with self-contained proofs, and identifies open problems. arXiv:2607.05652

**arXiv ID:** 2205.15409 **Authors:** Aapo Hyvärinen **Published:** 2022-05-27T07:32:33Z **Abstract:** This book uses the modern theory of artificial intelligence (AI) to understand human suffering or mental pain. Both humans and sophisticated AI agents process information about the world in order to achieve goals and obtain rewards, which is why AI can be used as a model of the human brain and mind. This book intends to make the theory accessible to a relatively general audience, requiring o...

**arXiv ID:** 2407.01647 **Authors:** Parviz Ghafariasl, Masoomeh Zeinalnezhad, Amir Ahmadishokooh **Published:** 2024-07-01T05:24:19Z **Abstract:** Timely alerts about hazardous air pollutants are crucial for public health. However, existing forecasting models often overlook key factors like baseline parameters and missing data, limiting their accuracy. This study introduces a hybrid approach to address these issues, focusing on forecasting hourly PM2.5 concentrations using Support Vector Re...

Online Generalised Predictive Coding via Dynamic Expectation Maximisation (ODEM) for biologically plausible online learning. Activation: predictive coding, online learning, DEM, dynamic expectation maximisation, active inference.

On-Policy Distillation (OPD) methodology for transforming autoregressive models into diffusion language models efficiently, eliminating train-inference mismatch.

Nonstabilizerness diffusion dynamics methodology for analyzing magic resource generation in many-body quantum systems using stabilizer Renyi entropy and tensor network methods.