KIT-Bibliothek

05: Maschinelles Lernen 2 - Fortgeschrittene Verfahren, Vorlesung, SS 2017, 07.07.2017

This audio or video file is copyrighted. Access is only allowed via computers of the Karlsruhe Institute of Technology (KIT).

Author

Jacques Kaiser

Editor

KIT | Webcast

Participating institute

Institut für Anthropomatik und Robotik (IAR)

Genre

Vorlesung

Description

  • 0:00:00 Starten
  • 0:00:21 Motivation Video: SPAUN
  • 0:03:27 Overview
  • 0:04:27 Neurons anatomy
  • 0:05:35 Neurons in action
  • 0:07:42 Modelling Neurons
  • 0:09:04 Synaptic plasticity
  • 0:11:41 Coding information
  • 0:14:06 Why not backpropagation in spiking networks?
  • 0:17:05 Os backpropagation biologically possible
  • 0:21:43 Synaptic plasticity as learning
  • 0:23:41 Different types of long-term synaptic plasticity
  • 0:24:18 Formalization of precise spiketime plasticity rules
  • 0:27:23 Example: Hebbian and anti-hebbian learning rules
  • 0:28:55 Formalization of rate-based plasticity rules
  • 0:32:56 Formalization of reward-based plasticity rules
  • 0:34:50 Other neuromodulators
  • 0:36:32 Structural plasticity - rewiring synapses
  • 0:40:14 Unsupervised learning with STDP
  • 0:42:43 Supervised learning with STDP - Associative learning
  • 0:44:18 SPORE - Synaptic Plasticity with Online Reinforcement learning
  • 0:45:45 SPORE in action - Binary classification
  • 0:47:08 Spiking Networks as kernel methods
  • 0:48:44 Linear reression on rates - Neural Engineeering Framework
  • 0:52:13 Regression on Post-Synaptic Potentials - Liquid State Machines
  • 0:54:07 Liquid State Machines in real life
  • 0:56:21 Neuromorphic chips - spiking network
  • 1:01:52 Neuromorphic sensors -Silicon retina (DVS)
  • 1:03:54 Short-term visual prediction of address events
  • 1:04:36 Connecting robots to spiking networks
  • 1:05:41 Grasping motions with spiking networks
  • 1:06:37 End-to-end robot control - Braitenberg vehicles
  • 1:09:44 mBuilding Braitenberg vehicles in simulation
  • 1:11:53 Where to o next?

Duration (hh:mm:ss)

01:12:49

Series

Maschinelles Lernen 2 - Fortgeschrittene Verfahren, Vorlesung, SS 2017

Published on

11.07.2017

Subject area

Computer science

License

KITopen Licence

Resolution 1280 x 720 Pixel
Aspect ratio 16:9
Audio bitrate 112661 bps
Audio channels 2
Audio Codec aac
Audio Sample Rate 48000 Hz
Total Bitrate 917900 bps
Color Space yuv420p
Container mov,mp4,m4a,3gp,3g2,mj2
Media Type video/mp4
Duration 4370 s
Filename DIVA-2017-413_hd.mp4
File Size 4.096 byte
Frame Rate 25
Video Bitrate 799143 bps
Video Codec h264

Embed Code