Sleep prevents catastrophic forgetting in spiking neural networks by forming a joint synaptic weight representation.
Sleep prevents catastrophic forgetting in spiking neural networks by forming a joint synaptic weight representation.
复制标题
DOI：
10.1371/journal.pcbi.1010628
复制
发表时间：
2022-11
影响因子：
4.3
通讯作者：
中科院分区：
文献类型：
--
作者：
人工神经网络在顺序训练时会覆盖先前学习的任务，这种现象被称为灾难性遗忘。相比之下，大脑不断学习，并且通常在新的训练与巩固记忆的睡眠期交织时学习得最好。在这里，我们使用尖峰网络来研究灾难性遗忘背后的机制以及睡眠在防止它方面的作用。该网络可以被训练来学习复杂的觅食任务，但当在不同的任务上连续训练时，表现出灾难性遗忘。在突触权重空间中，新任务训练使突触权重配置远离代表旧任务的流形，导致遗忘。将新任务训练与离线重新激活的周期交织，模仿生物睡眠，通过将网络突触权重状态约束到先前学习的流形来减轻灾难性遗忘，同时允许权重配置朝向表示旧任务和新任务的流形的交叉点收敛。这项研究揭示了大脑在睡眠期间应用突触权重动态的可能策略，以防止遗忘和优化学习。人工神经网络可以在许多领域实现超人的性能。尽管取得了这些进步，但这些网络在顺序学习中失败了;它们在新任务上实现了最佳性能，而牺牲了先前学习任务的性能。另一方面，人类和动物具有持续学习并将新数据纳入现有知识库的非凡能力。睡眠被认为在记忆和学习中起着重要的作用，它使先前学习的记忆模式能够自发地重新激活。在这里，我们使用一个尖峰神经网络模型，模拟动物大脑中的感觉处理和强化学习，以证明将新任务训练与睡眠样活动交织在一起优化了网络在突触权重空间中的记忆表示，以防止忘记旧记忆。睡眠通过重放旧的记忆痕迹而不显式地使用旧的任务数据来实现这一点。
Artificial neural networks overwrite previously learned tasks when trained sequentially, a phenomenon known as catastrophic forgetting. In contrast, the brain learns continuously, and typically learns best when new training is interleaved with periods of sleep for memory consolidation. Here we used spiking network to study mechanisms behind catastrophic forgetting and the role of sleep in preventing it. The network could be trained to learn a complex foraging task but exhibited catastrophic forgetting when trained sequentially on different tasks. In synaptic weight space, new task training moved the synaptic weight configuration away from the manifold representing old task leading to forgetting. Interleaving new task training with periods of off-line reactivation, mimicking biological sleep, mitigated catastrophic forgetting by constraining the network synaptic weight state to the previously learned manifold, while allowing the weight configuration to converge towards the intersection of the manifolds representing old and new tasks. The study reveals a possible strategy of synaptic weights dynamics the brain applies during sleep to prevent forgetting and optimize learning. Artificial neural networks can achieve superhuman performance in many domains. Despite these advances, these networks fail in sequential learning; they achieve optimal performance on newer tasks at the expense of performance on previously learned tasks. Humans and animals on the other hand have a remarkable ability to learn continuously and incorporate new data into their corpus of existing knowledge. Sleep has been hypothesized to play an important role in memory and learning by enabling spontaneous reactivation of previously learned memory patterns. Here we use a spiking neural network model, simulating sensory processing and reinforcement learning in animal brain, to demonstrate that interleaving new task training with sleep-like activity optimizes the network’s memory representation in synaptic weight space to prevent forgetting old memories. Sleep makes this possible by replaying old memory traces without the explicit usage of the old task data.
登录
查看更多内容
影响因子：
2.5
作者：
Farries, Michael A.;Fairhall, Adrienne L.
通讯作者：
Fairhall, Adrienne L.
影响因子：
25
作者：
Ji, Daoyun;Wilson, Matthew A.
通讯作者：
Wilson, Matthew A.
影响因子：
16.2
作者：
Bazhenov M;Stopfer M
通讯作者：
Stopfer M
影响因子：
2.4
作者：
Bazhenov, M;Rulkov, NF;Timofeev, I
通讯作者：
Timofeev, I
影响因子：
2.9
作者：
Hayes, Tyler L.;Krishnan, Giri P.;Bazhenov, Maxim;Siegelmann, Hava T.;Sejnowski, Terrence J.;Kanan, Christopher
通讯作者：
Kanan, Christopher