Skip to content

Latest

arXiv cs.NE Papers

NeuroSynth: A Biologically Inspired Continual Reinforcement Learning Architecture for Mitigating Catastrophic Forgetting

arXiv:2607.28663v1 Announce Type: new Abstract: Artificial Intelligence (AI) systems often perform well on isolated tasks but struggle under continual learning conditions, where training on new tasks…

arXiv cs.NE Papers

Information Processing by Neuron Populations in the Central Nervous System: A Theory of the Mathematical Structure of Data and Operations

arXiv:2309.02332v3 Announce Type: replace-cross Abstract: In the mammalian central nervous system, neurons are organized into populations communicating by spike trains propagating along axonal bundles. How such…

arXiv cs.NE Papers

Hash Chemistry: Minimal Models for Evolutionary Growth of Complexity

arXiv:2607.28219v1 Announce Type: cross Abstract: Hash Chemistry is a family of minimalistic evolutionary models in which a deterministic hash function assigns a scalar score to entities…

arXiv cs.NE Papers

Automatic programming via large language models with population self-evolution for dynamic fuzzy job shop scheduling problem

arXiv:2410.22657v2 Announce Type: replace Abstract: Heuristic dispatching rules (HDRs) are widely used for solving the dynamic fuzzy job shop scheduling problem (DFJSSP). However, their performance is…

arXiv cs.NE Papers

The Fast Lane Hypothesis: Von Economo Neurons Implement a Biological Speed-Accuracy Tradeoff

arXiv:2604.09229v2 Announce Type: replace Abstract: von Economo neurons (VENs) are large bipolar projection neurons found exclusively in the anterior cingulate cortex (ACC) and frontal insula of…

arXiv cs.NE Papers

Neural Architecture Search for Traffic Prediction: A Survey of Methods, Challenges, and Future Directions

arXiv:2607.26467v1 Announce Type: cross Abstract: Traffic prediction is a core task in intelligent transportation systems, supporting applications such as adaptive signal control, route guidance, and ride-hailing…

arXiv cs.NE Papers

EvoPINN: Agentic Discovery of Executable Algorithms for Physics-Informed Neural Networks

arXiv:2607.26490v1 Announce Type: cross Abstract: Physics-informed neural networks (PINNs) have emerged as a powerful paradigm for solving partial differential equations (PDEs), yet their performance heavily relies…

arXiv cs.NE Papers

Emergent Sparsity in Frozen Random CNN Feature Extractors for Deep Reinforcement Learning

arXiv:2607.26059v1 Announce Type: cross Abstract: We report a striking phenomenon: deep reinforcement learning agents trained with frozen, randomly initialized CNN feature extractors spontaneously develop extremely sparse…

arXiv cs.NE Papers

The Sparsity Ceiling: Where Spiking Networks Can and Cannot Trade Activity for Energy

arXiv:2607.26648v1 Announce Type: new Abstract: Spiking neural networks (SNNs) are promoted as an energy-efficient substrate because sparse, event-driven activity replaces dense multiply-accumulates with cheap accumulates. We…

arXiv cs.NE Papers

Shared Symbolic Backbones for Physically Consistent Multi-Output Symbolic Regression

arXiv:2607.26528v1 Announce Type: new Abstract: Symbolic regression provides analytical expressions, but it is usually applied one output at a time. This is limiting in process systems,…