05 2021 档案

Reinforcement Learning in Continuous Time and Space
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! Neural Computation, no. 1 (2000): 219-245 Abstract 本文提出了一种用于连续时间动态系统的强化学习框架,而没有时间、状态和动作的离散先验。基于哈密尔顿-雅各比-贝尔曼(HJB)方程的无穷视野 阅读全文

posted @ 2021-05-19 15:09 穷酸秀才大草包 阅读(461) 评论(0) 推荐(0)

A Learning Theory for Reward-Modulated Spike-Timing-Dependent Plasticity with Application to Biofeedback
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! PLoS Computational Biology, no. 10 (2008): e1000180-e1000180 阅读全文

posted @ 2021-05-12 22:05 穷酸秀才大草包 阅读(131) 评论(0) 推荐(0)

Functional Requirements for Reward-Modulated Spike-Timing-Dependent Plasticity
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! The Journal of neuroscience: the official journal of the Society for Neuroscience, no. 40 (2010): 13326-13337 Abstract 阅读全文

posted @ 2021-05-12 21:29 穷酸秀才大草包 阅读(279) 评论(0) 推荐(0)

BindsNET学习系列 ——Reward
摘要:相关源码:bindsnet/bindsnet/learning/reward.py 1、AbstractReward class AbstractReward(ABC): # language=rst """ Abstract base class for reward computation. " 阅读全文

posted @ 2021-05-11 15:05 穷酸秀才大草包 阅读(183) 评论(0) 推荐(0)

Reinforcement Learning Through Modulation of Spike-Timing-Dependent Synaptic Plasticity
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! Neural Computation, no. 6 (2007): 1468-1502 Abstract 突触效应的持续改变(作为突触前和突触后相对时间的函数关系)是一种被称为脉冲时序依赖可塑性(STDP)的现象。在此,我们显示了通过全局 阅读全文

posted @ 2021-05-07 16:58 穷酸秀才大草包 阅读(715) 评论(0) 推荐(0)

BindsNET MNIST源码分析
摘要:BindsNET:https://github.com/BindsNET/bindsnet 相关代码:bindsnet/examples/mnist/ 1、eth_mnist.py: import os import torch import argparse import numpy as np 阅读全文

posted @ 2021-05-06 15:57 穷酸秀才大草包 阅读(257) 评论(0) 推荐(0)

The Dance of the Interneurons: How Inhibition Facilitates Fast Compressible and Reversible Learning in Hippocampus
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! bioRxiv, pp.318303-47, (2018) Abstract 海马能够快速学习传入的信息,即使该信息仅被观察到一次也是如此。此外,在"尖峰波纹"(SPW-R)期间,能够以前向或反向模式以压缩格式回放此信息。我们利用最先进的 阅读全文

posted @ 2021-05-06 11:18 穷酸秀才大草包 阅读(112) 评论(0) 推荐(0)

Replay in biological and artificial neural networks
摘要:原文参见DeepMind的博客:https://www.deepmind.com/blog/article/replay-in-biological-and-artificial-neural-networks 作者:Zeb Kurth-Nelson, Will Dabney One of a se 阅读全文

posted @ 2021-05-05 11:38 穷酸秀才大草包 阅读(310) 评论(0) 推荐(0)

Fine-Tuning and the Stability of Recurrent Neural Networks
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! PloS one, no. 9 (2011): e22885-e22885 Abstract 对构建稳定循环模型网络的标准理论方法的主要批评是,需要对突触连接权重进行微调。这种批评之所以严峻,是因为有关学习这些权重所提出的规则已显示出对其 阅读全文

posted @ 2021-05-03 10:19 穷酸秀才大草包 阅读(184) 评论(0) 推荐(0)

Adaptive Critics and the Basal Ganglia
摘要:郑重声明:原文参见标题,如有侵权,请联系作者,将会撤销发布! This Article is brought to you for free and open access by the Computer Science at ScholarWorks@UMass Amherst. It has b 阅读全文

posted @ 2021-05-03 10:19 穷酸秀才大草包 阅读(220) 评论(0) 推荐(0)

导航