Data & Analytics

**arXiv ID:** 1911.10735 **Authors:** Julien Girard-Satabin, Guillaume Charpiat, Zakaria Chihani, Marc Schoenauer **Published:** 2019-11-25T07:28:45Z **Abstract:** The topic of provable deep neural network robustness has raised considerable interest in recent years. Most research has focused on adversarial robustness, which studies the robustness of perceptive models in the neighbourhood of particular samples. However, other works have proved global properties of smaller neural networks. Yet,...

CaMBRAIN methodology for real-time continuous EEG inference using causal Mamba state space models. First model enabling long-range streaming inference of variable-length EEG signals with >10x higher throughput.

**arXiv ID:** 2303.10761 **Authors:** Ruslan Vasilev, Alexander D'yakonov **Published:** 2023-03-19T20:27:51Z **Abstract:** Neural networks solving real-world problems are often required not only to make accurate predictions but also to provide a confidence level in the forecast. The calibration of a model indicates how close the estimated confidence is to the true probability. This paper presents a survey of confidence calibration problems in the context of neural networks and provides an em...

Calibrated measurement framework using the Brody exponent β as a quantitative measure of short-range exclusion in 2D spatial point processes. Originally from quantum chaos level-spacing statistics, now calibrated for spatial analysis with corrected CSR baseline, empirical β-r_excl calibration (Spearman ρ=0.988), and control protocols. Use for quantum chaos analysis, spatial statistics, prime number embeddings, and manufactured surface characterization.

**arXiv ID:** 2501.03264 **Authors:** Qi Wang, Marco Federici, Herke van Hoof **Published:** 2025-01-04T03:28:21Z **Abstract:** The neural process (NP) is a family of computationally efficient models for learning distributions over functions. However, it suffers from under-fitting and shows suboptimal performance in practice. Researchers have primarily focused on incorporating diverse structural inductive biases, \textit{e.g.} attention or convolution, in modeling. The topic of inference subo...

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

BrainDyn: A Sheaf Neural ODE framework for modeling continuous-time brain dynamics on structured graphs. Combines LSTM stalks with sheaf Laplacian message passing and neural ODE evolution. Apply when: brain dynamics modeling, fMRI/EEG forecasting, generative brain models, neural ODEs, sheaf theory, brain graph networks, perturbation prediction, synthetic brain data. Keywords: sheaf neural ODE, brain dynamics, fMRI modeling, EEG forecasting, brain graphs, sheaf Laplacian, neural ODE, brain reg...

BrainCast methodology for spatio-temporal forecasting of whole-brain fMRI time series. Uses dual-branch architecture with ST-CausalConv for spatial decoding and ST-Mixer for temporal prediction. Activation: fMRI forecasting, brain time series prediction, spatio-temporal brain modeling.

Brain-LLM alignment is driven by training-language dominance, not an inherent property of English. Tests with fMRI from 112 participants across English, Chinese, French and 7 LLMs (English-dominant, Chinese-dominant, multilingual). Baichuan2-7B reverses alignment gradient entirely; typological distance independently affects alignment degradation in syntax regions (IFG). Accepted at CoNLL 2026. Activation: brain-LLM alignment, cross-linguistic brain encoding, training data dominance, multiling...

Graph Neural Network methods for brain connectivity analysis. Use when analyzing fMRI/EEG brain network data, modeling brain structure-function relationships, predicting cognitive outcomes from connectome data, or applying GNN to neuroscience problems. Keywords: brain graph, connectome GNN, neural network brain, fMRI GNN, brain connectivity analysis, 脑网络图神经网络, 脑连接性分析, 认知预测.

RE-CONFIRM framework for validating robustness of biomarkers discovered by brain foundation models from dynamic functional connectivity. Systematic evaluation of internal reliability, external reliability, and validity for clinical biomarkers. Activation: RE-CONFIRM, biomarker validation, brain foundation model, robust biomarkers, dynamic functional connectivity.

Mathematical framework for quantifying the value of brain data for machine learning. Derives scaling laws, exchange rates between brain and task samples, and conditions for robustness gains via neural regularization. Activation: brain data value, neural data worth, brain-regularized learning, neuroai scaling laws, brain sample exchange rate.

Comparative methodology for brain alignment across learning rules (BP, FA, PC, STDP). Key finding: single training epoch reduces V1 alignment by 25-90%. BP most destructive, PC and STDP preserve brain-like structure. Use when: brain alignment, representational similarity analysis, biologically plausible learning, visual cortex modeling, learning rule comparison. arXiv: 2605.30556

**arXiv ID:** 2008.04213 **Authors:** Yuan Sun, Sheng Wang, Yunzhuang Shen, Xiaodong Li, Andreas T. Ernst, Michael Kirley **Published:** 2020-07-29T13:03:37Z **Abstract:** This paper introduces an enhanced meta-heuristic (ML-ACO) that combines machine learning (ML) and ant colony optimization (ACO) to solve combinatorial optimization problems. To illustrate the underlying mechanism of our ML-ACO algorithm, we start by describing a test problem, the orienteering problem. In this problem, the o...

Learning biophysical Hodgkin-Huxley models from extracellular MEA data for precise neurostimulation prediction

**arXiv ID:** 2408.05237 **Authors:** Akshansh Mishra **Published:** 2024-08-05T13:27:54Z **Abstract:** This study presents a novel approach to predicting mechanical properties of Additive Friction Stir Deposited (AFSD) aluminum alloy walled structures using biomimetic machine learning. The research combines numerical modeling of the AFSD process with genetic algorithm-optimized machine learning models to predict von Mises stress and logarithmic strain. Finite element analysis was employed to...

**arXiv ID:** 1706.04052 **Authors:** Jinzhuo Wang, Wenmin Wang, Ronggang Wang, Wen Gao **Published:** 2017-06-13T13:30:04Z **Abstract:** Monte Carlo tree search (MCTS) is extremely popular in computer Go which determines each action by enormous simulations in a broad and deep search tree. However, human experts select most actions by pattern analysis and careful evaluation rather than brute search of millions of future nteractions. In this paper, we propose a computer Go system that follows ...

Large-scale AI benchmarking methodology for cancer detection models. Evaluates tumor-detection AI across tumor size, location, demographic subgroups, and imaging protocols using 85,355 CT scans and 12 models. Use when: benchmarking medical AI models, evaluating cancer detection systems, assessing subgroup fairness in healthcare AI, analyzing CT scan AI performance, building robust tumor detection pipelines.

Behavior-dLDS: decomposed linear dynamical systems model for neural activity partially constrained by behavior. Disentangles behavior-related neural dynamics from internal computations in large-scale neural recordings. Scales to tens of thousands of neurons. Use when modeling neural population dynamics, decomposing brain activity into behavioral vs. internal subsystems, or analyzing brain-wide recordings with behavioral correlates. Activation: behavior-dLDS, decomposed linear dynamical system...

BCMI-driven motion control detection using EEG-based machine learning and interaction entropy for high-order brain networks during music-assisted driving

BCI-sift (BCI Systematic and Interpretable Feature Tuning) methodology for automated feature selection in Brain-Computer Interface applications. Integrates advanced optimization algorithms (scikit-learn compatible) to identify informative neural features across electrode, temporal, and frequency dimensions from HD ECoG and other BCI modalities. Activates on BCI feature selection, ECoG decoding optimization, neural feature tuning, automated BCI ML pipeline, brain-computer interface classificat...

Bayesian decision-making framework for membership inference attacks on statistical releases using Bayesian network population models. Reframes membership inference with respect to populations represented as Bayesian networks, enabling more effective specialized attacks by incorporating prior information about attribute dependency structures. Use when analyzing statistical disclosure risk, designing membership inference attacks, or evaluating privacy of released statistics.

Bayesian dynamical framework for modeling time-order effects in sequential haptic perception. Captures perceptual biases from prior expectations and temporal structure using drift-diffusion dynamics. Activation: haptic perception, Bayesian dynamics, time-order effects, sequential stimuli, perceptual bias.

**arXiv ID:** 2205.13881 **Authors:** Steven Adriaensen, André Biedenkapp, Gresa Shala, Noor Awad, Theresa Eimer, Marius Lindauer, Frank Hutter **Published:** 2022-05-27T10:30:25Z **Abstract:** The performance of an algorithm often critically depends on its parameter configuration. While a variety of automated algorithm configuration methods have been proposed to relieve users from the tedious and error-prone task of manually tuning parameters, there is still a lot of untapped potential as th...