Intelligence scales superlinearly with accessible dimensional freedom. By operating in ℝⁿ (n>3) spaces, cognitive systems achieve exponential gains in capacity, connectivity, and problem-solving abilities through dimensional shortcuts and hyper-connectivity unavailable in traditional 3D models.
Peer-reviewed papers and preprints on Neuro-Dimensional Architecture
This foundational paper introduces Neuro-Dimensional Architecture (NDA), proposing that cognitive systems operate in higher-dimensional mathematical spaces. It presents the Dimensional Intelligence Hypothesis (DIH) and establishes four foundational constructs for understanding higher-dimensional cognition.
This paper introduces Dimensional Resonance Binding (DRB), a theoretical framework within NDA that explains how the brain achieves flexible, context-dependent cognition through transient resonant alignment of task-relevant low-dimensional manifolds across neural substrates.
Detailed mathematical formulation of the Dimensional Intelligence Hypothesis (DIH), including formal proofs of dimensional scaling laws, experimental predictions for neuroscience research, and validation protocols for artificial systems.
Application of NDA principles to transformer architectures, demonstrating how attention mechanisms operating in higher-dimensional spaces achieve exponential improvements in context length, parameter efficiency, and reasoning capabilities.
Four foundational constructs for higher-dimensional cognition
Studies neural embeddings in ℝⁿ (n>3), enabling hyper-connectivity and distance compression in cognitive spaces.
Models cognitive processes propagating simultaneously across multiple independent dimensions, enabling parallel intuition.
Utilizes curved manifolds for exponential memory capacity and natural hierarchy embedding.
Explains decision-making and insight as projection from high-dimensional states to lower-dimensional outputs.
Open-source implementations of NDA architectures
Reference implementations are currently in development and will be released under the MIT License.
Hyper-Dimensional Transformer extending attention to n-dimensional token manifolds
Non-Euclidean Memory Network using hyperbolic geometry for exponential capacity
Learns task-optimal projections from high to low-dimensional spaces
Note: GitHub repository will be populated following paper publication.
Leading the development of the NDA framework
Founder & Principal Researcher
Author of the Neuro-Dimensional Architecture (NDA) framework and founder of the Neuro-Dimensional Architecture Initiative. With a background in cognitive science and artificial intelligence, Bilal leads research into higher-dimensional cognition and its applications in both biological and artificial systems.
Get in touch with the NDA Initiative
Neuro-Dimensional Architecture Initiative
A research organization advancing higher-dimensional cognition