Data & Analytics

Finite-temperature quantum Krylov method for computing thermal properties of quantum many-body systems from real-time overlaps. Use when analyzing quantum many-body systems at finite temperatures, computing thermal observables, or avoiding thermal state preparation in quantum simulations.

Fermi-Dirac quantization methodology for neural networks — reinterprets classical neurons as parameterized Hamiltonians and replaces variables with quantum operators. BQP-complete for certain decision problems. Use when: designing quantum neural architectures, quantizing activation functions (ReLU, GeLU, sigmoid), building hybrid quantum-classical neural algorithms, analyzing quantum advantage in neural computation, or studying the quantum-classical boundary in machine learning.

**arXiv ID:** 1710.09300 **Authors:** Filipe Alves Neto Verri, Renato Tinós, Liang Zhao **Published:** 2017-10-25T15:18:27Z **Abstract:** Data and knowledge representation are fundamental concepts in machine learning. The quality of the representation impacts the performance of the learning model directly. Feature learning transforms or enhances raw data to structures that are effectively exploited by those models. In recent years, several works have been using complex networks for data repre...

Dynamical Lie Algebra (DLA) framework for navigating the expressivity-trainability paradox in QML - using group-theoretic geometric priors as structural regularizers to guarantee scalable, gradient-rich training landscapes.

Extended predictive coding framework using exponential family distributions beyond Gaussian assumptions. Reveals biological neural network properties: nonlinearity, heterogeneity, biological plausibility. Maintains FEP-PC correspondence up to second cumulant. Derives biologically plausible local plasticity rules from EFD variational free energy. Use when: predictive coding, free energy principle, exponential family, variational inference, biological plausibility, local plasticity rules, neura...

Exploiting Symmetry in Quantum Reservoir Computing (QRC) methodology — observable-orbit completion aligns encoding, dynamics, measurement, and readout so symmetry-induced inductive bias is visible in the measured feature map; validated on spin-ring, real-weather cyclic forecasting, and IBM hardware.

Exclusion statistics as a thermodynamic resource in quantum heat engines — using particle statistics interpolation (fermion/boson/anyon) as a design parameter for quantum thermal machines. From arXiv:2606.19310.

Excitation-driven data generation and distributed control optimization for building thermal systems and district heating networks. Combines BuilDyn framework (arXiv:2605.29849) and distributed NMPC with ADMM (arXiv:2605.29841).

**arXiv ID:** 2402.00070 **Authors:** Yushu Jiang **Published:** 2024-01-30T19:37:21Z **Abstract:** Extensive fine-tuning on Large Language Models does not always yield better results. Oftentimes, models tend to get better at imitating one form of data without gaining greater reasoning ability and may even end up losing some intelligence. Here I introduce EvoMerge, a systematic approach to large language model training and merging. Leveraging model merging for weight crossover and fine-tuning...

**arXiv ID:** 2205.10116 **Authors:** Gabriel Wang, Anish Thite, Rodd Talebi, Anthony D'Achille, Alex Mussa, Jason Zutty **Published:** 2022-05-12T17:27:38Z **Abstract:** Machine Learning models are used in a wide variety of domains. However, machine learning methods often require a large amount of data in order to be successful. This is especially troublesome in domains where collecting real-world data is difficult and/or expensive. Data simulators do exist for many of these domains, but the...

**arXiv ID:** 2511.20909 **Authors:** Anil K. Saini, Jose Guadalupe Hernandez, Emily F. Wong, Debanshi Misra, Tiffani J. Bright, Jason H. Moore **Published:** 2025-11-25T22:50:59Z **Abstract:** Machine learning models trained on real-world data may inadvertently make biased predictions that negatively impact marginalized communities. Reweighting, which assigns a weight to each data point used during model training, can mitigate such bias, though sometimes at the cost of predictive accuracy. I...

**arXiv ID:** 1907.04482 **Authors:** Cheng He, Shihua Huang, Ran Cheng, Kay Chen Tan, Yaochu Jin **Published:** 2019-07-10T01:50:20Z **Abstract:** Recently, more and more works have proposed to drive evolutionary algorithms using machine learning models.Usually, the performance of such model based evolutionary algorithms is highly dependent on the training qualities of the adopted models.Since it usually requires a certain amount of data (i.e. the candidate solutions generated by the algorit...

**arXiv ID:** 2311.07485 **Authors:** Mohammad Mahdi Rahimi, Hasnain Irshad Bhatti, Younghyun Park, Humaira Kousar, Jaekyun Moon **Published:** 2023-11-13T17:25:06Z **Abstract:** Federated Learning (FL) is a decentralized machine learning paradigm that enables collaborative model training across dispersed nodes without having to force individual nodes to share data. However, its broad adoption is hindered by the high communication costs of transmitting a large number of model parameters. This...

**arXiv ID:** 2608.19888 **Authors:** Kentaro Oda **Published:** 2026-08-20T10:54:17Z **Abstract:** Streaming systems that maintain a pool of expert models must repeatedly decide whether to reuse an existing expert for arriving data, spawn a new one, or defer. We present a decision layer that makes all three outcomes statistically meaningful. Reuse and spawn are posed as one-sided sequential hypotheses on a conditional (mechanism-level) discrepancy, separated by an indifference zone; defer is...

**arXiv ID:** 1603.06212 **Authors:** Randal S. Olson, Nathan Bartley, Ryan J. Urbanowicz, Jason H. Moore **Published:** 2016-03-20T13:32:27Z **Abstract:** As the field of data science continues to grow, there will be an ever-increasing demand for tools that make machine learning accessible to non-experts. In this paper, we introduce the concept of tree-based pipeline optimization for automating one of the most tedious parts of machine learning---pipeline design. We implement an open source T...

**arXiv ID:** 2312.14681 **Authors:** Raffaele Marino, Lorenzo Buffoni, Lorenzo Chicchi, Lorenzo Giambagli, Duccio Fanelli **Published:** 2023-12-22T13:34:18Z **Abstract:** EODECA (Engineered Ordinary Differential Equations as Classification Algorithm) is a novel approach at the intersection of machine learning and dynamical systems theory, presenting a unique framework for classification tasks [1]. This method stands out with its dynamical system structure, utilizing ordinary differential eq...

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...

Embodied Virtual Reality feedback reshapes neural representations to support continuous 3D motor imagery decoding in brain-computer interfaces. First systematic investigation of embodied VR feedback during real-time 3D virtual limb control. Use when: (1) Designing VR-based BCI systems, (2) Studying motor imagery neural representations, (3) Comparing VR vs screen feedback modalities, (4) Investigating longitudinal BCI training effects. Activation: embodied VR feedback, motor imagery BCI, 3D vi...

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.

Open-source platform cataloguing 791 public neurophysiological datasets (EEG, MEG, iEEG, EMG, fNIRS) with automatic format repair, BIDS compliance, and machine learning integration.

EEG-based tinnitus biomarker identification methodology with cross-dataset generalization. Uses microstate analysis and Koopman operator analysis via DMD to extract robust neural signatures. Focuses on Koopman eigenvalue magnitude for oscillation stability. Applications: clinical diagnostics, cross-platform tinnitus detection. Triggers: tinnitus biomarker, EEG microstate, Koopman EEG, cross-dataset generalization

Deep learning framework for objective consciousness level measurement using multi-dimensional transcranial electrical stimulation (TES) with EEG. Combines TES-evoked brain responses with CNN classification for bedside-awareness assessment. Activation triggers: eeg tes, consciousness measurement, transcranial stimulation, brain state classification, awareness assessment, disorder of consciousness.

Structure-Guided Diffusion Model (SGDM v3) for EEG-based visual cognition reconstruction with enhanced cross-subject generalization. Combines structurally supervised VAE, spatiotemporal EEG encoder with contrastive learning, and ControlNet-guided diffusion for high-fidelity image reconstruction from brain signals.