メニュー
このリポジトリの skills
SkillsMP は hiyenwong/ai_collection から 4,114 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
hiyenwong/ai_collection収集済み skill 4,114 件中 40 件を表示しています。
Bayesian framework for Information Processing Pathway Maps (IPPMs) to map cortical entrainment from EEG/MEG data. Compares Bayesian vs frequentist approaches for model adjudication in computational neuroscience. Uses temporal response functions (TRFs) and…
原文の言語: 英語
Bayesian framework for quantifying neural entrainment evidence in Information Processing Pathway Maps (IPPMs). Replaces frequentist null hypothesis testing with probabilistic model adjudication using Bayes factors. Enables robust comparison of competing…
原文の言語: 英語
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…
原文の言語: 英語
Bayesian neural network methodology for robust portfolio management in dynamic financial markets. Uses Bayesian inference to quantify uncertainty in portfolio optimization, adapt to changing market conditions, and provide probabilistic risk assessments. Use…
原文の言語: 英語
Adversarial robustness methodology for EEG-based Brain-Computer Interfaces (BCIs). Lightweight custom CNN architectures that outperform EEGNet/DeepConvNet/SleepEEGNet under gradient-based adversarial attacks. Use for: BCI security, adversarial defense, EEG…
原文の言語: 英語
Optimized BCI rehabilitation protocols for stroke recovery. Addresses task design, training duration, and neuroplasticity-driven adaptation for maximizing post-stroke motor recovery through brain-computer interfaces.
原文の言語: 英語
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…
原文の言語: 英語
BCMI-driven motion control detection using EEG-based machine learning and interaction entropy for high-order brain networks during music-assisted driving
原文の言語: 英語
BEAST3D方法论：基于3D Gaussian splatting的自监督动物行为分析与神经编码框架。从稀疏视角视频重建3D行为结构，生成视角不变特征用于神经活动预测。
原文の言語: 中国語
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.…
原文の言語: 英語
Multi-level representational probing framework for evaluating digital twins of sensory cortex beyond standard prediction accuracy. Probes latent representations (linear decodability, latent-unit tuning, population geometry) in mouse V1 digital twins. Based on…
原文の言語: 英語
Brain-Inspired Capture (BI-Cap) — neuromimetic perceptual simulation for visual decoding from neural signals. Emulates Human Visual System processing with dynamic/static transformations and MI-guided blur regulation. Evidence-driven latent space handles…
原文の言語: 英語
Bilinear gating methodology linking dendritic coincidence detection to goal-directed adaptation. Motor cortex neurons encode goal information in burst fraction (not firing rate), implementing bilinear gate G(g)·Y(s) where goal and state inputs multiply via…
原文の言語: 英語
Brain-Inspired Mixture of Experts (BiMoE) framework for EEG-dominant affective state recognition. Uses brain-topology-aware expert partitioning with dual-stream encoders and adaptive routing for multimodal sentiment analysis combining EEG with peripheral…
原文の言語: 英語
Causal analysis of Binary Spiking Neural Networks (BSNNs) using logic-based explainable AI methods. Formally defines BSNNs as binary causal models and provides tractable algorithms for computing abductive explanations.
原文の言語: 英語
[TODO: Complete and informative explanation of what the skill does and when to use it. Include WHEN to use this skill - specific scenarios, file types, or tasks that trigger it.]
原文の言語: 英語
BioMysteryBench methodology for evaluating AI bioinformatics research capabilities. Tasks models with analyzing real-world biological datasets using method-agnostic, ground-truth-verified questions that allow superhuman evaluation.
原文の言語: 英語
生物神经元参数学习SNN方法论。联合优化突触权重和神经元内在参数，结合Lempel-Ziv复杂度实现可解释的时空脉冲数据分类。适用于脉冲神经网络、神经形态计算、可解释AI。触发词：生物神经元、SNN参数学习、Lempel-Ziv复杂度、神经形态计算、biological neuron、LZC、spiking neural network。
原文の言語: 中国語
Hybrid optimization combining quantum-behaved particle swarm optimization with bio-inspired swarm intelligence mechanisms. Integrates quantum-probabilistic position updates with ant colony and bee foraging behavior for multimodal and biologically structured…
原文の言語: 英語
Automated framework for assessing biological plausibility of spiking neuron models using Izhikevich firing pattern classification. Use when evaluating spiking neuron models, comparing neuromorphic designs, or quantifying how well artificial neurons replicate…
原文の言語: 英語
BioMysteryBench methodology for benchmarking LLM bioinformatics research capabilities on real-world datasets with consensus-based grading and path-independent evaluation
原文の言語: 英語
BiSpikCLM methodology — the first fully binary spiking MatMul-free causal language model. Integrates Softmax-Free Spiking Attention (SFSA) and Spike-Aware Alignment Distillation (SpAD) to train energy-efficient spiking LLMs. Use when building spiking language…
原文の言語: 英語
BLEG (LLM-Enhanced Brain Graph Analysis) methodology. Integrates LLMs with brain graph neural networks for improved neurological disease classification via knowledge-enhanced connectivity representation.
原文の言語: 英語
BLEG (Brain LLM Enhanced Graph) - using Large Language Models as fMRI graph enhancers for brain network analysis. LLM-augmented GNNs for sparse neurograph learning. Keywords: LLM, fMRI, brain network, GNN, graph enhancement, multimodal fusion.
原文の言語: 英語
行为指导的神经种群动力学建模方法论（BLEND）。通过特权知识蒸馏， 在训练时利用行为信号，推理时仅需神经活动。 触发词：神经种群动力学、行为建模、知识蒸馏、特权信息、BLEND、 neural population dynamics, behavior-guided, privileged distillation。
原文の言語: 複数言語
BlockRank methodology for scalable In-context Ranking (ICR) with LLMs. Enforces structured sparse attention (linear complexity) + auxiliary contrastive loss for efficient document ranking. Activation triggers: in-context ranking, blockrank, listwise…
原文の言語: 英語
Boltzmann Attention methodology — energy-based attention mechanism using learnable Ising model pairwise couplings for cooperative attention. Augments data-dependent local fields with learnable inter-position correlations, enabling diabatic quantum annealing…
原文の言語: 英語
Loss-tolerant quantum communication using Bosonic Gottesman-Kitaev-Preskill (GKP) parity encoding. Implements quantum repeaters with concatenated Bell state measurement for long-distance quantum communication. Activation: bosonic GKP, quantum repeater,…
原文の言語: 英語
Bosonic quantum error correction using Gottesman-Kitaev-Preskill (GKP) grid states and programmable nonlinear bosonic circuits. Covers grid state preparation, logical qubit encoding in continuous-variable oscillators, Wigner function characterization, and…
原文の言語: 英語
Approximability limits for bounded-degree max-LINSAT and implications for decoded quantum interferometry
原文の言語: 英語
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,…
原文の言語: 英語
Brain alignment of vision-language models (VLMs) and large-action models (LAMs) with fMRI during naturalistic gameplay. Use when: studying brain-AI alignment during interactive tasks, comparing VLMs vs LAMs neural encoding, analyzing action vs reasoning…
原文の言語: 英語
Generative models for brain "brainstorming" — studying the brain's spontaneous idea generation using free energy principle, critical dynamics, and default mode network analysis. Based on Smith et al. (2026) linking neural criticality, predictive coding, and…
原文の言語: 英語
Causal visual representation discovery framework for neuroscience. Use when analyzing brain region representations through causal testing rather than mere activation maximization. Covers counterfactual stimulus generation, image-to-fMRI encoding models,…
原文の言語: 英語
BrainCause methodology for discovering and causally validating visual representations in the human brain using generative models, counterfactual stimulus synthesis, and fMRI encoding models. Use when: (1) studying causal vs correlational brain…
原文の言語: 英語
Hierarchical organization of critical brain dynamics. Analyze how brain networks exhibit critical behavior across multiple scales, including neuronal avalanches, power-law distributions, and long-range temporal correlations. Use when studying brain…
原文の言語: 英語
Critical assessment methodology for evaluating the brain criticality hypothesis using rigorous statistical and computational approaches. Use for analyzing criticality in neural systems, avalanche dynamics, and evaluating claims of critical brain states.…
原文の言語: 英語
Critical assessment methodology for evaluating the brain criticality hypothesis. Proposes Memory-Induced Long-Range Order (MILRO) as an alternative explanation for scale-invariant correlations in neural activity. Use for analyzing neural avalanches,…
原文の言語: 英語
Memory-Induced Long-Range Order (MILRO) assessment framework challenging the brain criticality hypothesis. Analyzes scale-invariant correlations in neural activity as stable phase rather than critical point. Keywords: brain criticality, MILRO, scale-free,…
原文の言語: 英語
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…
原文の言語: 英語