Research & Publications

Evidence-Based Inquiry at the Frontier of Neural Computing

Our research program spans theoretical neuroscience, applied neuromorphic engineering, and neurotrophic computing architectures. All findings are subject to peer review and open scientific scrutiny.

Published Papers

Active Patents

Research Areas

Ongoing Studies

Publications

Neurotrophic ComputingComputational NeuroscienceJournal Article

2024

Neurotrophic Modulation of Synaptic Plasticity in Silicon Neural Networks

Chen, R., Vasquez, M., Okafor, T.

Nature Neuromorphic Systems — Vol. 3, Issue 2

We demonstrate that encoding BDNF-mediated signaling pathways as computational primitives in silicon neural networks yields adaptive learning systems with 10× improved energy efficiency over conventional deep learning architectures, without catastrophic forgetting.

DOI:10.1038/nns.2024.0031

Neuromorphic EngineeringHardwareJournal Article

2024

Spike-Timing-Dependent Plasticity at Scale: A Neuromorphic Hardware Implementation

Vasquez, M., Park, J., Chen, R.

IEEE Transactions on Neural Networks and Learning Systems — Vol. 35, Issue 4

This paper presents the NC-1 neuromorphic processing unit, implementing leaky integrate-and-fire neuron models with on-chip STDP learning rules. Benchmark results demonstrate sub-2.4ms end-to-end inference latency at 0.8W power draw.

DOI:10.1109/tnnls.2024.3012847

Neurotrophic ComputingComputational NeuroscienceJournal Article

2023

Topological Reconfiguration in Growth-Factor-Mediated Computational Graphs

Okafor, T., Lindqvist, S., Vasquez, M.

Neural Computation — Vol. 35, Issue 11

We introduce a formal model of neurotrophic-guided graph reconfiguration, proving convergence properties under biologically-plausible update rules. The model predicts network topology evolution with 94.2% accuracy on held-out biological datasets.

DOI:10.1162/neco_a_01623

Neuromorphic EngineeringHardwareSurvey

2023

Energy-Efficient Inference on Neuromorphic Substrates for Edge Deployment

Park, J., Chen, R., Okafor, T.

ACM Computing Surveys — Vol. 56, Issue 3

A systematic survey and empirical evaluation of neuromorphic inference pipelines for edge deployment scenarios. We benchmark seven architectures across latency, power, and accuracy, identifying the NC-1 substrate as Pareto-optimal for real-time classification tasks.

DOI:10.1145/3584371

Neurotrophic ComputingComputational NeuroscienceJournal Article

2022

First-Principles Derivation of Hebbian Learning from Neurotrophic Signaling Dynamics

Lindqvist, S., Chen, R.

PLOS Computational Biology — Vol. 18, Issue 9

We derive Hebbian learning rules directly from the kinetics of neurotrophin receptor binding and downstream MAPK/ERK signaling cascades, providing a biochemical grounding for classical associative learning in artificial systems.

DOI:10.1371/journal.pcbi.1010482

Neuromorphic EngineeringHardwareJournal Article

2022

Neuromorphic Sensor Fusion for Real-Time Tactile Processing

Vasquez, M., Park, J.

Advanced Intelligent Systems — Vol. 4, Issue 7

We present a neuromorphic pipeline for real-time tactile sensor fusion using event-driven spike encoding. The system achieves 0.3ms response latency with 97.8% classification accuracy on a standardized tactile benchmark dataset.

DOI:10.1002/aisy.202200041

Patents

GrantedFiled 2022

Adaptive Neurotrophic Computational Graph with Dynamic Topology Reconfiguration

US 11,847,392 B2

GrantedFiled 2021

On-Chip STDP Learning Engine for Neuromorphic Processing Units

US 11,623,018 B1

PendingFiled 2023

Event-Driven Spike Encoding Architecture for Low-Latency Sensor Fusion

US 2024/0089147 A1