Nature Neuroscience, Journal Year: 2014, Volume and Issue: 17(12), P. 1635 - 1643
Published: Nov. 17, 2014
Language: Английский
Nature Neuroscience, Journal Year: 2014, Volume and Issue: 17(12), P. 1635 - 1643
Published: Nov. 17, 2014
Language: Английский
Cell, Journal Year: 2015, Volume and Issue: 162(3), P. 622 - 634
Published: July 1, 2015
Language: Английский
Citations
975Nature reviews. Neuroscience, Journal Year: 2016, Volume and Issue: 17(3), P. 183 - 195
Published: Feb. 11, 2016
Language: Английский
Citations
882Cell, Journal Year: 2015, Volume and Issue: 162(3), P. 635 - 647
Published: July 1, 2015
Language: Английский
Citations
770Nature, Journal Year: 2016, Volume and Issue: 535(7613), P. 505 - 510
Published: July 1, 2016
Language: Английский
Citations
576Neuron, Journal Year: 2018, Volume and Issue: 101(1), P. 133 - 151.e7
Published: Nov. 29, 2018
Language: Английский
Citations
470Nature, Journal Year: 2021, Volume and Issue: 599(7886), P. 650 - 656
Published: Nov. 3, 2021
Language: Английский
Citations
394Nature Neuroscience, Journal Year: 2018, Volume and Issue: 21(9), P. 1260 - 1271
Published: Aug. 6, 2018
Language: Английский
Citations
364Neuron, Journal Year: 2018, Volume and Issue: 97(2), P. 434 - 449.e4
Published: Jan. 1, 2018
Language: Английский
Citations
358Nature Communications, Journal Year: 2020, Volume and Issue: 11(1)
Published: July 17, 2020
Abstract Recurrently connected networks of spiking neurons underlie the astounding information processing capabilities brain. Yet in spite extensive research, how they can learn through synaptic plasticity to carry out complex network computations remains unclear. We argue that two pieces this puzzle were provided by experimental data from neuroscience. A mathematical result tells us these need be combined enable biologically plausible online learning gradient descent, particular deep reinforcement learning. This method–called e-prop–approaches performance backpropagation time (BPTT), best-known method for training recurrent neural machine In addition, it suggests a powerful on-chip energy-efficient spike-based hardware artificial intelligence.
Language: Английский
Citations
357Neuron, Journal Year: 2015, Volume and Issue: 85(2), P. 429 - 438
Published: Jan. 1, 2015
Language: Английский
Citations
346