Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

Video course · 10 chapters · 27 min · certificate

Biological Learning Algorithms: Backpropagation vs Predictive Coding Explained

How might the brain learn? The credit assignment problem, why backpropagation is biologically implausible, predictive coding and its energy formalism, activity and weight update rules, the neural connectivity they need, and how the full system learns.

Become a memberAlready a member? Sign in

What you'll learn

  • Explain the credit assignment problem
  • Identify why backprop is biologically implausible
  • Describe predictive coding
  • Understand energy-based formulations
  • Derive local activity and weight updates
  • See how predictive coding learns as a whole

Chapters

10 chapters · 26:51
  1. 1:15 01Intro Members Introduction How does the brain learn?
  2. 1:34 02Concepts Members Credit assignment problem Which weights to change?
  3. 3:16 03Plausibility Members Problems with backprop Why the brain likely doesn't use it.
  4. 2:02 04Theory Members Foundations of predictive coding Predicting the senses.
  5. 3:01 05Math Members Energy formalism One number for the state.
  6. 4:04 06Dynamics Members Activity update rule Rolling downhill.
  7. 2:30 07Circuits Members Neural connectivity Where errors live.
  8. 3:16 08Learning Members Weight update rule Local synaptic change.
  9. 4:17 09System Members Putting it all together Learning from experience.
  10. 1:36 10Wrap-up Members Outro Further learning.