arXiv subjects

Mario Negrello

Publications and source records attributed to Mario Negrello.

4 recordsLinked to original sources

arXiv2026

The Neuroscience of Transformers

Neuroscience has long informed the development of artificial neural networks, but the success of modern architectures invites, in turn, the converse: can modern networks teach us lessons about brain function? Here, we examine the structure of the cortical column and propose that the transformer provides a natural computational analogy for multiple elements of cortical microcircuit organization. Rather than claiming a literal implementation of transformer equations in cortex, we develop a hypothetical mapping between transformer operations and laminar cortical features, using the analogy as an orienting framework for analysis and discussion. This mapping allows us to examine in greater depth how contextual selection, content routing, recurrent integration, and interlaminar transformations may be distributed across cortical circuitry. In doing so, we generate a broad set of predictions and experimentally testable hypotheses concerning laminar specialization, contextual modulation, dendritic integration, oscillatory coordination, and the effective connectivity of cortical columns. This proposal is intended as a structured hypothesis rather than a definitive account of cortical computation. Placing transformer operations and cortical architectonics into a common descriptive framework sharpens questions, reveals new functional correspondences, and opens a productive route for reciprocal exchange between systems neuroscience and modern AI. More broadly, this perspective suggests that comparing brains and architectures at the level of computational organization can yield genuine insight into both.

q-bio.NC↗

arXiv2025

Spatial Spiking Neural Networks Enable Efficient and Robust Temporal Computation

The efficiency of modern machine intelligence depends on high accuracy with minimal computational cost. In spiking neural networks (SNNs), synaptic delays are crucial for encoding temporal structure, yet existing models treat them as fully trainable, unconstrained parameters, leading to large memory footprints, higher computational demand, and a departure from biological plausibility. In the brain, however, delays arise from physical distances between neurons embedded in space. Building on this principle, we introduce Spatial Spiking Neural Networks (SpSNNs), a framework in which neurons learn coordinates in a finite-dimensional Euclidean space and delays emerge from inter-neuron distances. This replaces per-synapse delay learning with position learning, substantially reducing parameter count while retaining temporal expressiveness. Across the Yin-Yang and Spiking Heidelberg Digits benchmarks, SpSNNs outperform SNNs with unconstrained delays despite using far fewer parameters. Performance consistently peaks in 2D and 3D networks rather than infinite-dimensional delay spaces, revealing a geometric regularization effect. Moreover, dynamically sparsified SpSNNs maintain full accuracy even at 90% sparsity, matching standard delay-trained SNNs while using up to 18x fewer parameters. Because learned spatial layouts map naturally onto hardware geometries, SpSNNs lend themselves to efficient neuromorphic implementation. Methodologically, SpSNNs compute exact delay gradients via automatic differentiation with custom-derived rules, supporting arbitrary neuron models and architectures. Altogether, SpSNNs provide a principled platform for exploring spatial structure in temporal computation and offer a hardware-friendly substrate for scalable, energy-efficient neuromorphic intelligence.

cs.NE↗

arXiv2024

Gradient Diffusion: Sensitivity-Matrix Co-Simulation Enables Activity Adaptation and Learnable Plasticity in Neural Simulators

Computational neuroscience relies on large-scale dynamical-systems models of neurons, with a vast amount of offline, pre-simulation, tuned parameters, with models often tied to their brain simulators. These fixed parameters lead to stiff models, that show unnatural behaviour when introduced to new environments, or when combined into larger networks. In contrast to offline tuning, in biology, cells continuously adapt via homeostatic plasticity to stay in desired dynamical regimes. In this work, we aim to introduce such online tuning of cellular parameters into brain simulation. We show that the sensitivity equation of a biorealistic neural models has the same shape as a general neuron model, and can be simulated within existing brain simulators. Via co-simulation with the sensitivity equation, we enable both offline, and online tuning of activity of arbitrary biophysically realistic brain models. Furthermore, we show that this opens the possibility to study the biological mechanisms underlying homeostatic plasticity, via both meta-learning plasticity mechanism as well as treating online tuning as a black-box plasticity mechanism. Through the generality of our methods, we hope that more computational science fields can capitalize on the similarity between the simulated model and its gradient system.

q-bio.NC↗

arXiv2021

EDEN: A high-performance, general-purpose, NeuroML-based neural simulator

Modern neuroscience employs in silico experimentation on ever-increasing and more detailed neural networks. The high modelling detail goes hand in hand with the need for high model reproducibility, reusability and transparency. Besides, the size of the models and the long timescales under study mandate the use of a simulation system with high computational performance, so as to provide an acceptable time to result. In this work, we present EDEN (Extensible Dynamics Engine for Networks), a new general-purpose, NeuroML-based neural simulator that achieves both high model flexibility and high computational performance, through an innovative model-analysis and code-generation technique. The simulator runs NeuroML v2 models directly, eliminating the need for users to learn yet another simulator-specific, model-specification language. EDEN's functional correctness and computational performance were assessed through NeuroML models available on the NeuroML-DB and Open Source Brain model repositories. In qualitative experiments, the results produced by EDEN were verified against the established NEURON simulator, for a wide range of models. At the same time, computational-performance benchmarks reveal that EDEN runs up to 2 orders-of-magnitude faster than NEURON on a typical desktop computer, and does so without additional effort from the user. Finally, and without added user effort, EDEN has been built from scratch to scale seamlessly over multiple CPUs and across computer clusters, when available.

q-bio.NC↗