Design & UX

Covariance-based method with Granger-causality refinement for recovering sparse neural connectivity from partial measurements.

REVE (Representation for EEG with Versatile Embeddings) - EEG foundation model trained on 60,000 hours from 25,000 subjects with novel 4D positional encoding for arbitrary electrode configurations. Achieves SOTA on 10 downstream tasks. Activation triggers: EEG foundation model, REVE, versatile embeddings, 4D positional encoding, cross-dataset EEG, brain-computer interface.

Four-generation framework for quantum biomedical sensors based on quantum resource utilization. Covers clinical translation challenges, noise limits, ensemble vs single-particle sensing. Use when: quantum biosensors, biomedical quantum sensing, clinical quantum sensors, quantum medical imaging sensors, biosensor generations, quantum resource biosensing, NV center biosensing, atomic magnetometer biomedical, quantum optical biosensing, macroscopic vs microscopic quantum sensing.

Parametrically induced strong coupling between superconducting quantum circuits and solid-state spin ensembles. Uses parametric pump to achieve on-demand MHz-rate coupling for quantum state transfer. Enables hybrid quantum memories with coherence beyond superconducting circuits alone. Use when designing quantum memory interfaces, spin-circuit coupling, or parametric quantum interconnects.

OmniNeuro multimodal HCI framework for explainable BCI feedback — integrates Physics (Energy), Chaos (Fractal Complexity), and Quantum-Inspired uncertainty modeling to transform BCI from silent decoder to transparent feedback partner. Use when designing brain-computer interfaces with interpretability, neurofeedback systems, BCI sonification, multimodal BCI feedback, or quantum-inspired uncertainty in neural decoding (arXiv: 2601.00843)

Spatiotemporal TDANN for modeling self-organized MT direction selectivity maps in the dorsal stream. Uses 3D ResNet with Momentum Contrast (MoCo) self-supervised learning and biological spatial loss to produce direction-selective pinwheel structures matching macaque MT physiology. Use when modeling cortical topographic self-organization, dorsal stream computation, direction selectivity, or spatiotemporal contrastive learning for visual neuroscience. arXiv: 2605.11718 (q-bio.NC, cs.AI, cs.NE)....

利用MICrONS功能连接组学数据构建生物合理的RNN。整合皮质几何、解剖连接和功能关系作为归纳偏置，实现更有效的学习和收敛到生物计算的组织原则。

脑结构-功能耦合线性生成框架。从扩散加权成像(DWI)结构连接预测静息态fMRI功能连接，揭示整合枢纽和中介枢纽的机制，支持虚拟损伤实验。适用于脑连接组学、神经影像分析、结构-功能关系研究。触发词：结构功能耦合、脑连接组、结构连接、功能连接、整合枢纽、structure-function coupling、structural connectivity、functional connectivity、integrator hub、mediator hub。

Layer-wise Interactive Dual-Stream Network (LI-DSN) for EEG Motor Imagery decoding with cross-subject generalization, spatial-temporal dual-stream architecture, and layer-wise interactive fusion. Addresses the challenge of EEG signal variability across subjects through dual-stream processing and interactive fusion at each network layer. Use when: EEG motor imagery decoding, cross-subject EEG classification, dual-stream neural networks, layer-wise feature fusion, brain-computer interface class...

**arXiv ID:** 1308.2350 **Authors:** Jayanta K. Dutta, Bonny Banerjee **Published:** 2013-08-10T22:56:26Z **Abstract:** Learning features invariant to arbitrary transformations in the data is a requirement for any recognition system, biological or artificial. It is now widely accepted that simple cells in the primary visual cortex respond to features while the complex cells respond to features invariant to different transformations. We present a novel two-layered feedforward neural model that...

Framework for designing heterogeneous AI architectures that avoid the 'giant hippocampus' problem of applying one architectural template (like Transformers) to all cognitive tasks, instead using structurally diverse modules with standardized interfaces.

脑动力学生成模型综述框架。整合计算神经科学、非线性动力学、数据驱动方法的生成模型方法论，涵盖不同组织尺度和抽象层次。适用于脑动力学建模、神经数据分析、科学机器学习。触发词：脑动力学、生成模型、神经动力学、动态系统模型、brain dynamics、generative model、neural dynamics、computational neuroscience、Dynamical systems。

fMRI2GES: Dual brain decoding alignment framework for co-speech gesture reconstruction from fMRI signals. Maps brain responses to external stimuli and decodes gestural behavior through dual-alignment brain decoding. Activation: fMRI gesture reconstruction, brain-to-gesture decoding, fMRI2GES, co-speech gesture brain decoding, dual brain decoding alignment.

EyeBrain methodology for classifying left and right brain lateralization activity using pupil diameter and fixation duration. Non-invasive cognitive state detection via eye-tracking. Triggers: eyebrain, brain lateralization, pupil diameter, fixation duration, eye-tracking, cognitive load.

Embodied Virtual Reality feedback methodology for continuous 3D motor imagery BCI decoding. First systematic investigation showing VR feedback elicits more decodable and generalizable neural representations than screen feedback. CNN-LSTM decoder achieves r=0.762 under VR vs r=0.672 screen. Use when: (1) Designing continuous BCIs for intuitive motor control, (2) Implementing VR-based neurorehabilitation systems, (3) Studying neural representation generalization across feedback modalities, (4) ...

Dynamic neural manifolds methodology for flexible closed-loop control on neuromorphic hardware. Uses ring attractor networks with sensory-modulated control neurons (speed, shape, selection) to drive subspace rotations and fine-grained trajectory control in neural state space. Implemented on SpiNNaker 2 chip with robotic maze navigation validation.

Covariant quantum error correction methodology for quantum brain models. Evaluates CQEC purification protocols across radical-pair proteins with ab initio spin Hamiltonians, analyzing layer-specific coherence dynamics and T2 sensitivity.

Code-Modulated Motion Visual Evoked Potential (c-MVEP) methodology for brain-computer interfacing using motion stimulation instead of flickering. Use when: designing BCI paradigms, visual evoked potential stimulation, motion-based BCI, reducing visual fatigue in SSVEP/c-VEP systems, EEG-based BCI with pseudo-random sequences. Activation: c-MVEP, motion VEP BCI, code-modulated motion, visual evoked potential BCI, flicker-free BCI, SSVEP alternative, c-VEP alternative.

Framework for understanding the nonlinear scaling relationship between brain-computer interface bandwidth and meaningful human input/output capacity, distinguishing between raw bandwidth, decodable neural states, and information a person can actually use, confirm, and express through embodiment, learning, and subject expression constraints

Multi-objective optimization methodology for quantum computing workflows, combining compilation strategy selection, noise suppression, and error-mitigation. Based on QBalance framework (arXiv: 2605.02966) and action-space engineering for RL-based circuit routing. Use when: designing quantum compilation pipelines, optimizing NISQ device execution, selecting error-mitigation strategies, or formulating multi-objective quantum workflow problems.

**arXiv ID:** 1902.01722 **Authors:** Paavo Parmas **Published:** 2019-02-05T14:54:05Z **Abstract:** Backpropagation and the chain rule of derivatives have been prominent; however, the total derivative rule has not enjoyed the same amount of attention. In this work we show how the total derivative rule leads to an intuitive visual framework for creating gradient estimators on graphical models. In particular, previous "policy gradient theorems" are easily derived. We derive new gradient estima...

DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation (arXiv: 2608.20114)

Spatiotemporal TDANN for modeling self-organized MT direction selectivity maps in the dorsal stream. Uses 3D ResNet with Momentum Contrast (MoCo) self-supervised learning and biological spatial loss to produce direction-selective pinwheel structures matching macaque MT physiology. Use when modeling cortical topographic self-organization, dorsal stream computation, direction selectivity, or spatiotemporal contrastive learning for visual neuroscience. arXiv: 2605.11718 (q-bio.NC, cs.AI, cs.NE)....

利用MICrONS功能连接组学数据构建生物合理的RNN。整合皮质几何、解剖连接和功能关系作为归纳偏置，实现更有效的学习和收敛到生物计算的组织原则。