论文标题
通过学习记住来查找在线神经更新规则
Finding online neural update rules by learning to remember
论文作者
论文摘要
我们研究从头开始研究神经激活（身体）和权重（突触）的本地当地更新规则。我们代表每个重量的状态和小型向量激活，并使用（元）神经网络参数化其更新。不同的神经元类型由不同的嵌入向量表示，该向量允许所有神经元使用相同的两个功能。我们没有像在类似系统中那样直接使用进化或长期背部传播来直接培训目标，而是激励和研究一个不同的目标：记住过去的经验段。我们解释了该目标如何与标准的后传播培训和其他形式的学习形式有关。我们使用短期后传播来训练这个目标，并分析性能是不同网络类型和问题难度的函数。我们发现，这种分析对构成学习规则的内容提供了有趣的见解。我们还讨论了这种系统如何形成一种自然基材，以解决诸如情节记忆，元学习和辅助目标之类的主题。
We investigate learning of the online local update rules for neural activations (bodies) and weights (synapses) from scratch. We represent the states of each weight and activation by small vectors, and parameterize their updates using (meta-) neural networks. Different neuron types are represented by different embedding vectors which allows the same two functions to be used for all neurons. Instead of training directly for the objective using evolution or long term back-propagation, as is commonly done in similar systems, we motivate and study a different objective: That of remembering past snippets of experience. We explain how this objective relates to standard back-propagation training and other forms of learning. We train for this objective using short term back-propagation and analyze the performance as a function of both the different network types and the difficulty of the problem. We find that this analysis gives interesting insights onto what constitutes a learning rule. We also discuss how such system could form a natural substrate for addressing topics such as episodic memories, meta-learning and auxiliary objectives.
