archive
Every paper Pith has read. Search by title, abstract, or pith.
781 papers in cs.NE · page 1
-
Bounded gate stabilizes quantum fast-weight programmers
Stable Self-Modulating Quantum Fast-Weight Programmers with Bounded Memory Gates
-
Q-learning speeds genetic bin-packing solver by 50x
Hybridizing a Grouping Metaheuristic with Reinforcement Learning for the One-Dimensional Bin Packing Problem
-
Single dendritic compartment embeds LMS for in-context learning
Dendritic In-Context Learning in a Single-Layer Spiking Neural Network
-
Stacking ensemble flags early Alzheimer's from ADNI records
Predicting Early Stages Of Alzheimer's Disease And Identifying Key Biomarkers Using Deep Artificial Neural Network And Ensemble Of Machine Learning Methodologies
-
Phase-locked loop yields simple bursting neuron circuit
Electronic Bursting Neuron: design, equations and hardware implementation
-
Evolving WFC inputs improves quality for local domains
Evolutionary Wave Function Collapse
-
Metabolic optimizer loop stays bounded under mild resource rules
Mechanism and Stability Analysis of Metabolic Closed-Loop Metaheuristics
-
Metabolic optimizer jointly selects features and tunes classifiers
MMAO-Cls: Metabolic Multi-Agent Optimization for Joint Feature Selection and Classifier Tuning
-
Simple random mutations find self-replicators as easily as paired interactions
BFF: Simple explanations for complex phenomena
-
Metabolic optimizer reaches 28.07 mean offline error in dynamic tests
MMAO-Dyn: A Metabolic Multi-Agent Optimizer for Dynamic Optimization
-
MFEA-CoD finds diverse novel solutions across tasks with less redundancy
From Consistency to Collaborative Discovery: MFEA-CoD for Multitask Novelty Search
-
MFEA-CoD makes novelty search tasks collaborate across problems
From Consistency to Collaborative Discovery: MFEA-CoD for Multitask Novelty Search
-
Memristive signed weights sustain anti-phase attractors autonomously
Self-Organized Learning in Oscillatory Neural Networks with Memristive Signed Couplings
-
Evolved Transformers match or beat baseline on time series tasks
EVOTS: Evolutionary Transformer Search for Time Series Forecasting
-
GP symbolic regression results independent of population start method
Evaluation of Population Initialization Methods for Genetic Programming-based Symbolic Regression
-
D-HTM issues warnings 8 samples before local anomalies
Distributed Hierarchical Temporal Memory with Shared Associative Memory for Cross-Entity Preemptive Warning
-
Dual-stream networks learn representations under Dale's principle
Diffusing Blame: Task-Dependent Credit Assignment in Biologically Plausible Dual-Stream Networks
-
MMAO outperforms baselines on CEC2017 and TSPLIB benchmarks
A Large-Scale Empirical Evaluation of MMAO Under Fair-Budget Continuous and Discrete Benchmarks
-
Black-box measure bounds neural net error under input noise
Robustness of neural networks to random noise perturbations of their inputs
-
PGDS labels decision variables as drivers or blockers in 10-objective problems
Partition-Guided Distance Saliency: Bridging Decision and Objective Spaces in Many-Objective Optimization
-
Genetic algorithm equals clipped gradient descent with Hessian-controlled noise
Why can genetic algorithms work in high-dimensional search spaces?
-
Perspective mapping reduces integral R2 to weighted hypervolume
Computing the Integral R2 Indicator by Perspective Mapping and Box Decomposition
-
Mapping reduces integral R2 to weighted hypervolume
Computing the Integral R2 Indicator by Perspective Mapping and Box Decomposition
-
Box decompositions compute integral R2 in O(n log n) for 2-3 objectives
Computing the Integral R2 Indicator by Perspective Mapping and Box Decomposition
-
Metabolic loop drives adaptive metaheuristic search across domains
Minimal MMAO: A Resource-Closed-Loop Framework for Adaptive Metaheuristic Search
-
Slow self-credit builds durable behavioral residue after buffer removal
From Detecting Agency to Doing Work: Self-Caused Credit Builds a Durable Behavioral Self in a Minimal Spiking Agent
-
STABLE co-evolves configs and components using semantic models
Semantics-Aware Bilevel Co-Evolution: Towards Automated Multicomponent Algorithm Design
-
Evolution strategy finds much smaller CNNs for steering prediction
Evolutionary Hyperparameter Optimization to Find Lightweight CNN Models for Autonomous Steering
-
Geometric stability of neural codes tracks behavior apart from drift
Geometric Stability of Neural Population Codes: Regional Variation, Behavioral Relevance, and Circuit Dependence
-
Opposing waves train deep nets with local Hebbian rules
Supervised Hebbian learning in Deep Counterstream Associative Networks
-
Agents invent languages cutting LLM reasoning tokens 3-6x
When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning
-
KG traces lift travel LLM accuracy from 22% to 82%
Travel-Oriented Reasoning Large Language Model via Domain-Specific Knowledge Graphs
-
Complex neuron merges signal strength with timed spikes
Unified Complex-valued Neural Network: A Magnitude-Phase Computational Model for Event-Driven Neuromorphic Learning
-
Neuromorphic chip beats CPU on energy for most graph searches
Road to scalability for efficient graph search on massively parallel neuromorphic hardware
-
Closed-form gradient steers networks toward flat minima
Closed-Form Steepest Descent Direction toward Flat Minima: Reducing Upper Bounds on the Loss Hessian Eigenspectrum in Neural Networks
-
Variance dynamics model sets bat algorithm parameter bounds
Analysis of Parameter Settings for the Bat Algorithm Using Variance Evolution
-
Spiking Q-network cuts DBS charge 80% while suppressing oscillations 45%
Neuromorphic Energy-Aware Learning for Adaptive Deep Brain Stimulation
-
Weighted sums miss non-supported points on concave Pareto fronts
Comparing Scalar Objective Functions for Multi-Criteria Engineering Optimization
-
Metabolic loop enables endogenous adaptation in multi-agent optimizer
MMAO: A Metabolic Multi-Agent Optimizer with Endogenous Resource Allocation for Continuous and Discrete Optimization
-
Analog KANs with pruning cut area by 55% and power by 50%
Co-Optimization of Analog Kolmogorov-Arnold Networks for Low-Power Function Approximation in Flexible Electronics
-
SNN pruning reaches 95.6% accuracy at 90% sparsity
Criticality-Constrained Iterative Pruning for Energy-Efficient Spiking Neural Networks via Combined Importance Scoring
-
Toolkit unifies CANN simulation and attractor analysis
CANNs: A Toolkit for Research on Continuous Attractor Neural Networks
-
Late local search boosts constrained DE by 5.58 percent in U-score
DE-2LS: Differential Evolution with Lightweight Late Local Search for Constrained Numerical Optimization
-
Late local search raises DE optimization scores by 11%
DE-2LS: Differential Evolution with Late-Stage local-search for Unconstrained Single-Objective Numerical Optimization
-
Fourier evolution separates scaffold and substructure edits in molecules
Multi-Objective Molecular Generation with Frequency-Controlled Evolutionary Dynamics
-
Key-axis erase supplies content signals to recurrent gates
CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention
-
Key-axis erase validates recurrent chunk solver and adds free content gating
CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention
-
Output-use feedback builds organized networks in agent systems
Surviving by Serving: Functional Relevance Drives Self-Organization in Complex Adaptive Systems
-
Bézier curves steer random walks for competitive optimization
Random Walk on B\'ezier Curves for Global Optimization
-
R2 subset selection is NP-hard in 3 objectives but greedy achieves 1-1/e approx
Three-Objective Integral R2 Subset Selection: NP-Hardness and Submodular Approximation