arXiv:2607. 15217v1 Announce Type: cross Abstract: We present NeuronSoup, a neural computation architecture that replaces synchronous layer-by-layer processing with asynchronous, delay-mediated signal propagation through a pool of shared neurons.
arXiv:2405. 02369v2 Announce Type: replace-cross Abstract: In the past decade, many successful networks are on novel architectures, which almost exclusively use the same type of neurons.
arXiv:2608. 06963v1 Announce Type: new Abstract: Biologically plausible learning models aim to explain how neural circuits can implement effective learning under the constraints of real neurons.
Purin is a biology-inspired mechanism for artificial neural networks that adds synaptic efficacy modulation to conventional convolutional neural networks. It introduces a time‑interval‑based abstraction for neural activities, allowing short‑ and long‑term synaptic efficacy changes without discrete time‑steps. Experiments on AlexNet, VGG11, and GoogLeNet show that Purin improves classification accuracies across evaluated datasets.
arXiv:2608. 04358v1 Announce Type: new Abstract: Continual learning (CL) requires models to learn tasks sequentially, yet deep neural networks often suffer from plasticity loss and poor knowledge transfer, which can impede their long-term adaptability.
arXiv:2605. 08022v2 Announce Type: replace-cross Abstract: Spiking Neural Networks (SNNs) have been proposed as biologically plausible and energy-efficient alternatives to conventional Artificial Neural Networks (ANNs).
The paper compares genetic algorithm (GA) and gradient descent (GD) training for a distance‑encoding biomorphic‑informational neural network (DEBI‑NN) designed for low‑data medical datasets. A spatial backpropagation scheme was implemented for GD, and both optimizers were evaluated on synthetic, radiomic, and fetal cardiotocography datasets. Across all experiments, GA consistently outperformed GD, achieving higher classification accuracy and more stable decision boundaries, while GD struggled with the interdependent spatial parameters of DEBI‑NN.
arXiv:2605. 30370v2 Announce Type: replace-cross Abstract: From their inception in the 1950s, artificial neural networks (ANNs) started using the so-called point neuron model then prevalent in neuroscience, hoping that this analogy would allow for a better emulation of brain function.
NeuroSketch presents a practical design recipe for neural decoding, beginning with a comparative study of nine basic architectures that identifies CNN‑2D as the most effective. The recipe incorporates macro‑level gradual feature‑map expansion and early downsampling, along with micro‑level grouped convolutions, resulting in two variants—NeuroSketch‑Base (1.4M parameters) and NeuroSketch‑Large (4.2M parameters). Across nearly 5,000 experiments on eight tasks involving visual, auditory, and speech modalities and EEG, SEEG, and ECoG signals, both variants outperform ten baseline models on every task.
The paper introduces TNLearn, an open‑source Python package that automates the construction and training of task‑based neurons and networks. It argues that different tasks benefit from customized neurons that incorporate task‑specific prior knowledge, representing a shift from traditional task‑based architectures. The package, documented with technical exposition, API reference, and examples, is available on GitHub and integrated into the PyTorch ecosystem.
arXiv:2607. 16295v1 Announce Type: cross Abstract: Mechanistic interpretability has made significant strides in understanding neural network representations, with sparse dictionary learning (SDL) methods, most prominently sparse autoencoders, as a central paradigm.
arXiv:2607. 08406v1 Announce Type: new Abstract: Backpropagation (BP) dominates deep learning training, but its reliance on gradients brings inherent troubles -- vanishing and exploding gradients.