0% Complete
صفحه اصلی
/
سی و سومین کنفرانس بین المللی مهندسی برق
Wake-Sleep Learning in R-STDP-Based Spiking Neural Networks to Avoid Catastrophic Forgetting
نویسندگان :
Mehrdad Baradaran
1
Katayoon Kobraei
2
Saeed Reza Kheradpisheh
3
1- دانشگاه شهید بهشتی تهران
2- دانشگاه شهید بهشتی تهران
3- دانشگاه شهید بهشتی تهران
کلمات کلیدی :
Neuroscience-inspired AI،Spiking Neural Networks (SNNs)،Continual Learning،Catastrophic Forgetting،Generative Replay
چکیده :
Incremental Learning (IL) enables systems to learn new tasks over time without forgetting previous knowledge, a challenge known as catastrophic forgetting. This is especially prominent in Class-Incremental Learning (CIL), where models must sequentially learn new classes without task identity information. To address this problem, we propose a two-phase generative replay mechanism inspired by biological memory consolidation processes. During the "Day Phase," the model learns new tasks using reward-modulated Spike-Timing-Dependent Plasticity (R-STDP), while the "Night Phase" reinforces memory by generating and replaying synthetic data that simulates previously learned tasks. We evaluate our approach on Spiking Neural Networks (SNNs), leveraging their biologically inspired dynamics and energy-efficient architecture. Experiments were conducted on MNIST and Fashion-MNIST (FMNIST) datasets across three approaches: generative replay, original data replay, and a hybrid method combining both strategies. Experimental results demonstrate up to a 24% improvement in accuracy for CIL tasks on the MNIST dataset using the generative replay method, compared to standard SNNs without the Night Phase, highlighting the role of the Night Phase in enhancing performance. This work emphasizes the capability of SNNs to integrate biological principles into AI, bridging the gap between artificial intelligence and neuroscience.
لیست مقالات
لیست مقالات بایگانی شده
مقایسهگر پویا با قابلیت کار در شرایط زیر آستانه بر اساس منطق Pseudo-NMOS
سید سعید حسینی دولت آبادی - محسن جلالی
Optimal Path Planning of Mobile Robots using IsoCost-Based Dynamic Programming
Fatemeh Alvankarian - Ahmad Kalhor - Mehdi Tale Masouleh
Optimal Scheduling of Active Distribution Networks with High Penetration of Plug-in Electric vehicles and Renewables Using Grasshopper Optimization Algorithm
Seyyed Hadi Mousavi - Varahram Janatifar - Arya Abdolahi - Mitra Sarhangzadeh
Fusion of Multi-Level CNN With LBP Features For Facial Emotion Recognition
Ehsan Bahmanabady - Maryam Imani - Hassan Ghassemian
A New Protocol to Improve Effect of repetitive Transcranial Magnetic Stimulation in Treatment of Alzheimer's Disease
Ali Abedi - Gholamreza Moradi - Reza Sarraf Shirazi - Mehran Jahed
A New 1-D Model for Singular 2-D Systems
Kamyar Azarakhsh - Masoud Shafiee
Ultrahigh Step-Up Non-Isolated DC-DC Converter Based on Quadratic Converter without Coupled Inductor
Sajad Rostami - Vahid Abbasi - Masoumeh Parastesh
A modified Dempster Shafer approach to classification in surgical skill assessment
Arash Iranfar - Mohammad Soleymannejad - Behzad Moshiri - Hamid D. Taghirad
Improving Wind Turbines Blades Damage detection by using YOLO BoF and BoS
Reza Mohammadi - Saeed Sharifian
Energy Management of a Floating Photovoltaic System by Optimally Designed Fuzzy Logic Controller
Payam Loloei - S.hossein Hosseinian
بیشتر
ثمین همایش، سامانه مدیریت کنفرانس ها و جشنواره ها - نگارش 41.7.4