Jürgen Schmidhuber

Person on 3 sites

Deep-learning researcher at KAUST and IDSIA; co-invented the LSTM, and keeps a long public record of who published which idea first.

2 channels on ppll (opens ppll.app)
- x.com: @SchmidhuberAI (opens x.com)
  “Our embodied AI draws on memory. It recalls lived experiences, which contain depths of meaning that enable the AI to model the world, to plan, to minimize expected pain & maximize expected reward, and connect more…”
  
  2026-10-10
- Site: people.idsia.ch/~juergen (opens people.idsia.ch)
  His page: papers, the AI history posts and the priority disputes.
2 beliefs on korrents (opens korrents.com)
- Recurrent backpropagation learns slowly over long intervals mainly because error backflow decays. (opens korrents.com)
  “Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow.”
  
  1997-11-01
- LSTM can solve complex artificial long-time-lag tasks that prior recurrent network algorithms never solved. (opens korrents.com)
  “LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.”
  
  1997-11-01
1 recommendation on rcmnd (opens rcmnd.app)
- A New Kind of Science (opens rcmnd.app)
  “In fact Wolfram's book adds nothing substantial to Zuse's ideas, besides the pretty illustrations. Nor does it offer any predictions that go beyond Zuse's.”
  
  2026-10-24, book, by Stephen Wolfram, disliked

2 channels on ppll (opens ppll.app)