The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

Upside-Down Reinforcement Learning with Jürgen Schmidhuber - #357

Mar 16, 2020
Jürgen Schmidhuber, Co-founder of NNAISENSE and a pioneer in AI known for developing the LSTM network, discusses groundbreaking advancements in AI. He delves into Upside-Down Reinforcement Learning, highlighting its application in controlling robotic hands and improving learning through experiential methods. The conversation covers the shift from passive to active learning in industrial AI technologies, such as soft robotics and drone inspections, showcasing how curiosity-driven approaches can revolutionize robotics and real-world applications.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
ANECDOTE

NNAISENSE's Industrial AI Applications

  • NNAISENSE worked with Festo to build brains for robots controlled by air muscles.
  • They also partnered with EOS to improve industrial 3D printing processes using AI.
ANECDOTE

Self-Learning Toy Audi

  • NNAISENSE used reinforcement learning to teach a toy Audi to park.
  • The car used sensors and maximized pleasure signals while minimizing pain signals.
ANECDOTE

AI in Glassmaking

  • NNAISENSE is working with Schott, a glass manufacturer, to improve their glassmaking process.
  • They aim to control an industrial pipeline with molten glass at 1200 degrees Celsius.
Get the Snipd Podcast app to discover more snips from this episode
Get the app