Neural & Evolutionary Computing
Research on neural network architectures, evolutionary algorithms, and bio-inspired computing.
cs.NE · 188 papersGeometric analysis of attractor boundaries and storage capacity limits in kernel Hopfield networks
This paper investigates KLR Hopfield networks, revealing high storage capacity (P/N ~20) and that dynamical stability, not geometry, limits their ultimate storage.
NeuroRing: Scaling Spiking Neural Networks via Multi-FPGA Bidirectional Ring Topologies and Stream-Dataflow Architectures
NeuroRing is a scalable multi-FPGA SNN accelerator using a stream-dataflow and bidirectional ring topology for efficient, faster-than-real-time simulation.
Attractor FCM
Attractor FCM is a novel gradient-descent, physics-constrained Fuzzy Cognitive Map using residual memory, BPTT, and a fixed-point anchor for efficient learning.
Physical Foundation Models: Fixed hardware implementations of large-scale neural networks
Physical Foundation Models (PFMs) propose fixed hardware for large neural networks, leveraging physical dynamics for extreme efficiency and scale.
When Does Structure Matter in Continual Learning? Dimensionality Controls When Modularity Shapes Representational Geometry
This paper shows representational dimensionality dictates when modular architectures benefit continual learning, especially in low-dimensional regimes.
UniBCI: Towards a Unified Pretrained Model for Invasive Brain-Computer Interfaces
UniBCI is a unified pretrained model for invasive Brain-Computer Interfaces, achieving state-of-the-art performance and generalization across diverse tasks.
RCMAES: A Robust CMA-ES Variant for CEC2026 Competition
RCMAES is a new CMA-ES variant that uses adaptive population size reduction and restart for robust performance on CEC benchmarks.
Learning to Forget: Continual Learning with Adaptive Weight Decay
FADE introduces adaptive per-parameter weight decay for continual learning, improving knowledge retention and capacity management.
Causal Learning with Neural Assemblies
This paper shows neural assemblies can learn causal direction using a local plasticity mechanism, DIRECT, offering an auditable and explainable framework.
Population Dynamics in ARIEL Robotics Systems Featuring Embodied Evolution via Spatial Mating Mechanisms
This paper explores how spatial structure impacts evolutionary dynamics in robot populations using a spatially embedded evolutionary algorithm.
NORACL: Neurogenesis for Oracle-free Resource-Adaptive Continual Learning
NORACL uses neurogenesis to adaptively grow neural networks for continual learning, solving the oracle architecture problem and improving stability-plasticity.
Evolutionary feature selection for spiking neural network pattern classifiers
This paper extends evolutionary feature selection to JASTAP spiking neural networks, enabling smaller, more robust classifiers for noisy data.
Text-Utilization for Encoder-dominated Speech Recognition Models
This paper improves speech recognition by efficiently using text-only data in encoder-dominated models, showing simpler methods can be more effective.
Compressing ACAS-Xu Lookup Tables with Binary Decision Diagrams
This paper uses Binary Decision Diagrams (BDDs) to compress ACAS-Xu lookup tables, maintaining exact decision logic for verifiable, embedded deployment.
EdgeSpike: Spiking Neural Networks for Low-Power Autonomous Sensing in Edge IoT Architectures
EdgeSpike is a low-power spiking neural network framework for autonomous sensing in edge IoT, achieving high accuracy with significant energy savings.
EvoTSC: Evolving Feature Learning Models for Time Series Classification via Genetic Programming
EvoTSC uses genetic programming to automatically evolve lightweight and generalizable feature learning models for time series classification.
Benchmarking Stopping Criteria for Evolutionary Multi-objective Optimization
This paper introduces a new performance measure, a file-based benchmarking approach, and a data representation method for EMO stopping criteria.
The Effects of Population Size on the Performance of BEAGLE GPU-Based Genetic Programming Runs
This paper investigates how population size impacts GPU-based Genetic Programming performance for symbolic regression, finding varied optimal strategies.
Deployment-Aligned Low-Precision Neural Architecture Search for Spaceborne Edge AI
This paper introduces a hardware-aware NAS framework that integrates deployment-aligned low-precision training to improve accuracy on edge AI devices.
SeaEvo: Advancing Algorithm Discovery with Strategy Space Evolution
SeaEvo improves LLM-guided algorithm discovery by using explicit natural-language strategy descriptions to organize and guide evolutionary search.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.