2小时前更新- 01An Introduction and Tutorial of the Beagle FrameworkarXiv:2607.06731v1 Announce Type: new Abstract: The Beagle framework is a GPU-based genetic programming framework that enables highly efficient genetic programming search using large population sizes by leveraging NVIDIA GPUs. This technical guide provides an introduction to the Beagle framework and provides detailed instructions for using the framework for symbolic regression problems.Ilya Basin, Nathan Haut, Wolfgang Banzhaf
- 02Dynamic neural manifolds for flexible closed-loop control on neuromorphic hardwarearXiv:2607.07373v1 Announce Type: new Abstract: In biological circuits, sequential neural activity evolves along dynamic, low-dimensional manifolds to enable flexible behavior. Spiking network models link aspects of this sequential activity to features of manifold geometry through specific circuit mechanisms, making dynamic neural manifolds parameterizable, and thereby offering an explainable framework for neural computation. Extending this framework to neuromorphic engineering, we present an imOskar von Seeler, Christian Tetzlaff, Andrew Lehr
- 03Single-Entity Spiking Neuron Models: SurveyarXiv:2607.07429v1 Announce Type: new Abstract: In this work, we reviewed different approaches in mathematical modeling of biologically plausible neural systems. Models are characterized and classified based on their common features and special use cases. In addition to spiking models, different types of discrete and continuous analogs are considered to accurately simulate biological processes, including membrane potential dynamics. The models under investigation include neurons and various compLeon Parepko, Danila Shulepin, Albert Nasybullin
- 04Social-spatial dependencies for learning visual navigationarXiv:2607.07460v1 Announce Type: new Abstract: Navigation for social organisms rarely is a fully independent activity. Group structure and dynamics, as well as embodied interactions, critically influence useful behavior. Individual neural network controlled agents are trained to navigate in different social contexts, where social dependence and behavioral strategy learned is determined by relative task performance and spatial effect. Increasing high quality social information drives phase transPatrick Govoni, Pawel Romanczuk
- 05Intrinsic-Noise Consolidation: A Doob-Barrier-Conditioned Diffusion Turns Analog Device Noise into a Continual-Learning ResourcearXiv:2607.06924v1 Announce Type: cross Abstract: On analog neuromorphic hardware, intrinsic device noise is normally an accuracy tax. We ask whether it can instead consolidate memories. We cast per-synapse consolidation as a Doob h-transform: condition each weight's stochastic dynamics on never crossing a memory-critical barrier around its consolidated value. The conditioned diffusion gains an extra drift sigma^2 d/dw log h, a restoring force amplified by the noise variance itself that divergesGunner Levi Howe
- 06MEGO: Learning Mixture-of-Experts for General-Purpose Binary OptimizationarXiv:2405.18884v2 Announce Type: replace Abstract: Discrete optimization is ubiquitous in science and engineering. The vast array of existing discrete optimization problems, coupled with the continuous emergence of new ones, necessitates off-the-shelf optimizers capable of generating high-quality solutions for a large variety of optimization problems. This article introduces MEGO, a novel general-purpose neural optimizer for binary optimization under the black-box setting, intended for broad apShengcai Liu, Zhiyuan Wang, Yew-Soon Ong, Xin Yao, Ke Tang
- 07Optimization-Embedded Active Multi-Fidelity Surrogate Learning for Multi-Condition Airfoil Shape OptimizationarXiv:2603.17057v2 Announce Type: replace-cross Abstract: Active multi-fidelity surrogate modeling is developed for multi-condition airfoil shape optimization to reduce high-fidelity CFD cost while retaining RANS-consistent aerodynamic metrics. The framework couples a low-fidelity-informed Gaussian process regression transfer model with uncertainty-triggered sampling and a synchronized elitism rule embedded in a hybrid genetic algorithm. Low-fidelity XFOIL evaluations provide inexpensive featureIsaac Robledo, Alberto Vilari\~no, Arnau Mir\'o, Oriol Lehmkuhl, Carlos Sanmiguel Vila, Rodrigo Castellanos