1.088B 파라미터 Pure SNN의 Random Init 기반 수렴 성공
I scaled a pure Spiking Neural Network (SNN) to 1.088B parameters from scratch. Ran out of budget, but here is what I found
Preference Modeling 기반 Decision Transformer를 통한 행성 탐사 로버의 정렬 최적화
Human-Aligned Decision Transformers for planetary geology survey missions for low-power autonomous deployments
R-STDP 도입을 통한 Multi-Agent SNN의 탐색률 302% 향상 및 역할 유연성 확보
A Cognitive Neuroscience Study in Multi-Agent Box-Pushing Adversarial Games