A meta-learning approach to (re)discover plasticity rules that carve a desired function into a neural network
A meta-learning approach to (re)discover plasticity rules that carve a desired function into a neural network
复制标题
DOI：
10.1101/2020.10.24.353409
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Basile Confavreux;Everton J. Agnes;Friedemann Zenke;T. Lillicrap;T. Vogels
中科院分区：
文献类型：
--
作者：
Basile Confavreux;Everton J. Agnes;Friedemann Zenke;T. Lillicrap;T. Vogels
对生物学上忠实的突触可塑性规则的研究已经产生了大量的模型。它们通常受到实验数据的启发，并与之相适应，但它们很少产生服务于复杂功能的神经动力学。这些失败表明，目前的塑性模型仍然受到现有数据的约束。在这里，我们提出了一种替代方法，使用元学习来发现合理的突触可塑性规则。这些规则并不受实验数据的约束，而是受它们实现的功能和它们要产生的结构的约束。简而言之，我们通过沃尔泰拉扩展来参数化突触可塑性规则，然后使用监督学习方法（梯度下降或进化策略）来最小化问题相关的损失函数，该损失函数量化候选可塑性规则如何有效地将初始随机网络转换为具有所需功能的网络。我们首先通过重新发现先前描述的可塑性规则来验证我们的方法，从单神经元水平和“Oja规则”开始，这是一种简单的赫布可塑性规则，它捕获了神经元输入的大多数可变性的方向（即，第一主成分）。我们将问题扩展到网络级别，并要求框架找到Oja规则和反Hebbian规则，使得初始随机的双层发射率网络在学习后将恢复输入空间的几个主成分。接下来，我们将讨论具有可塑性抑制传入的整合和激发神经元网络。我们训练规则，通过对抗调谐激发来实现目标激发率。我们的算法发现了一个特定的子集的规则，可以解决这个任务的流形。我们的工作是一个自动化和公正的方法来揭示突触可塑性规则，服从生物学约束，可以解决复杂的功能的原则证明。
The search for biologically faithful synaptic plasticity rules has resulted in a large body of models. They are usually inspired by – and fitted to – experimental data, but they rarely produce neural dynamics that serve complex functions. These failures suggest that current plasticity models are still under-constrained by existing data. Here, we present an alternative approach that uses meta-learning to discover plausible synaptic plasticity rules. Instead of experimental data, the rules are constrained by the functions they implement and the structure they are meant to produce. Briefly, we parameterize synaptic plasticity rules by a Volterra expansion and then use supervised learning methods (gradient descent or evolutionary strategies) to minimize a problem-dependent loss function that quantifies how effectively a candidate plasticity rule transforms an initially random network into one with the desired function. We first validate our approach by re-discovering previously described plasticity rules, starting at the single-neuron level and “Oja’s rule”, a simple Hebbian plasticity rule that captures the direction of most variability of inputs to a neuron (i.e., the first principal component). We expand the problem to the network level and ask the framework to find Oja’s rule together with an anti-Hebbian rule such that an initially random two-layer firing-rate network will recover several principal components of the input space after learning. Next, we move to networks of integrate-and-fire neurons with plastic inhibitory afferents. We train for rules that achieve a target firing rate by countering tuned excitation. Our algorithm discovers a specific subset of the manifold of rules that can solve this task. Our work is a proof of principle of an automated and unbiased approach to unveil synaptic plasticity rules that obey biological constraints and can solve complex functions.