Department of Connectomics, Max Planck Institute for Brain Research, Max-von-Laue-Str. 4, 60438 Frankfurt, Germany.
Neuron. 2015 Apr 8;86(1):25-8. doi: 10.1016/j.neuron.2015.03.031.
Neuroscientists are generating data sets of enormous size, which are matching the complexity of real-world classification tasks. Machine learning has helped data analysis enormously but is often not as accurate as human data analysis. Here, Helmstaedter discusses the challenges and promises of neuroscience-inspired machine learning that lie ahead.
神经科学家正在生成规模巨大的数据集,这些数据集与现实世界分类任务的复杂性相匹配。机器学习极大地帮助了数据分析,但通常不如人类数据分析准确。在这里,Helmstaedter 讨论了神经科学启发的机器学习所面临的挑战和前景。