★★★★★ Jeff Hawkins (Thousand Brains Theory) • Machine Learning Street Talk #59

In this AI video ...

The ultimate goal of neuroscience is to learn how the human brain gives rise to human intelligence and what it means to be intelligent. Understanding how the brain works is considered one of humanity’s greatest challenges.

Jeff Hawkins thinks that the reality we perceive is a kind of simulation, a hallucination, a confabulation. He thinks that our brains are a model reality based on thousands of information streams originating from the sensors in our body. Critically – Hawkins doesn’t think there is just one model but rather; thousands.

Jeff has just released his new book, A thousand brains: a new theory of intelligence. It’s an inspiring and well-written book and I hope after watching this show; you will be inspired to read it too.

This video is relevant to everyone interested in artificial intelligence and machine learning.

Pod version: https://anchor.fm/machinelearningstreettalk/episodes/59—Jeff-Hawkins-Thousand-Brains-Theory-e16sb64

https://numenta.com/a-thousand-brains-by-jeff-hawkins
https://numenta.com/blog/2019/01/16/the-thousand-brains-theory-of-intelligence
https://numenta.com/assets/pdf/research-publications/papers/Sparsity-Enables-50x-Performance-Acceleration-Deep-Learning-Networks.pdf
https://numenta.com/neuroscience-research/research-publications/papers/a-framework-for-intelligence-and-cortical-function-based-on-grid-cells-in-the-neocortex

Your Brain Is Not an Onion With a Tiny Reptile Inside
https://journals.sagepub.com/doi/full/10.1177/0963721420917687

Pruning Neural Networks at Initialization: Why are We Missing the Mark?
https://arxiv.org/abs/2009.08576

Panel:
Dr. Keith Duggar https://twitter.com/DoctorDuggar
Connor Leahy https://twitter.com/npcollapse

Our thanks to:
Numenta
Matthieu Thiboust (https://www.insightsfromthebrain.com + https://twitter.com/mthiboust)
Shwetha Bharadwaj (show research https://www.linkedin.com/in/shwetha-bharadwaj-2b926a1b2)
Andreas Koepf (show research https://twitter.com/neurosp1ke)

Video Timeline

00:00 - Introduction
03:03 - The Neocortex
09:58 - Triune Brain
12:24 - Grid and place cells
14:54 - Reference frames
21:03 - Mountcastle
25:46 - Thousand brains theory of intelligence
32:40 - HTM
41:12 - Sparsity
52:57 - Main show kick off
54:36 - Tribalism in the ML Community
57:14 - Variation in approaches to the same goal
59:43 - Hawkins ideas validated, cortical uniformity
1:02:25 - Sparse distributed representations (SDRs)
1:06:08 - Reference frames as generalization
1:10:29 - Reference frame remapping
1:14:14 - Reference frames can generalize beyond three dimensions
1:17:26 - And generalize beyond spatial topology
1:20:12 - Intuitions behind why SDRs work well
1:24:03 - Are their capacity concerns with the SDR model
1:27:11 - At what level between GOFAI and Connectionism should focus our effort?
1:31:33 - The brain reasons by abstract movement through reference frames
1:35:34 - Human's don't know Universal Truth (if there is even such a thing)
1:37:34 - Learning elsewhere in the brain besides the neocortex
1:40:44 - Stochastic backpropagation in the human brain
1:44:04 - What's missing from artificial neural networks? Numenta's roadmap
1:48:59 - AGI Risk - the alignment problem
1:54:07 - AGI risk - the neocortex can thwart the old brain
1:57:47 - AGI risk - artificial evolution
2:01:18 - AGI risk - yes we need to think on and develop adequate control systems
2:03:48 - A balance of knowledge: innate, experiential, taught, or deduced
2:16:09 - Post-show wrap-up
2:16:59 - Advancements in direction at Numenta
2:19:50 - AGI risk recap
2:23:56 - Ought did evolve from Is, humans are the proof
2:26:29 - When AGI overcomes our weaknesses
2:28:54 - Who doesn't like forking?!
2:30:29 - Coherent synchronization as a measure of identity

YouTube Source for this AI Video

AI video(s) you might be interested in …