On-Line Learning of a Time-Dependent Rule
On-Line Learning of a Time-Dependent Rule
复制标题
时间相关规则的在线学习
DOI：
10.1209/0295-5075/20/8/012
复制
发表时间：
1992
期刊：
影响因子：
--
通讯作者：
H. Schwarze
中科院分区：
文献类型：
--
作者：
Michael Biehl;H. Schwarze
研究了神经网络中时间依赖线性可分规则的学习问题。该规则由执行随机游走的N向量表示。一个单层感知器的训练在线使用一个额外的权重衰减的Hebb类算法。在热力学极限N → ∞时，精确计算了推广误差的演化。我们考虑两种情况，随机抽取的训练示例和使用查询策略。规则永远不会被完美地学习，但可以在一定的误差水平内跟踪。仿真验证了分析结果。
We study the learning of a time-dependent linearly separable rule in a neural network. The rule is represented by an N-vector performing a random walk. A single-layer perceptron is trained on-line using a Hebb-like algorithm with an additional weight decay. The evolution of the generalization error is calculated exactly in the thermodynamic limit N → ∞. We consider both, training examples which are drawn randomly and using a query strategy. The rule is never learnt perfectly, but can be tracked within a certain error level. Simulations confirm the analytic results.