Steve Jurvetson: Deep Learning

black_boxbeyond_engineeringDeep Learning: Intelligence from Big Data
Sep 16, 2014
Stanford Graduate School of Business

A machine learning approach inspired by the human brain, Deep Learning is taking many industries by storm. Empowered by the latest generation of commodity computing, Deep Learning begins to derive significant value from Big Data. It has already radically improved the computer’s ability to recognize speech and identify objects in images, two fundamental hallmarks of human intelligence.

Industry giants such as Google, Facebook, and Baidu have acquired most of the dominant players in this space to improve their product offerings. At the same time, startup entrepreneurs are creating a new paradigm, Intelligence as a Service, by providing APIs that democratize access to Deep Learning algorithms.

Steve Jurvetson, Partner, DFJ Ventures

Adam Berenzweig, Co-founder and CTO, Clarifai
Naveen Rao, Co-founder and CEO, Nervana Systems
Elliot Turner, Founder and CEO, AlchemyAPI
Ilya Sutskever, Research Scientist, Google Brain

19:24 Subsystem inscrutability. Black box.

20:04 Beyond Engineering
Danny Hillis, The Pattern on the Stone: “The greatest achievement of our technology may well be the creation of tools that allow us to go beyond engineering–that allow us to create more than we can understand”

33:40 Convolutional Neural Networds

34:10 current deep learning model: recognizes 10,000 categories (classes).
average adult: 90,000 words (in English, not counting personal names)

1:09:00 we can propagate a signal with almost 0 loss. Biology can’t.
Will AI look like human intelligence? The underlying structure is different (substrate: silicon vs. neurons). With constraints, AI’s behavior will look however we want.


state of the art in reverse image search (01/01/2015):
the original (and only) source of the following image is placed low in results:

Unconscious determinants of free decisions (2008)

Unconscious determinants of free decisions in the human brain
Nature Neuroscience 11, 543 – 545 (2008)
Chun Siong Soon, … John-Dylan Haynes

There has been a long controversy as to whether subjectively ‘free’ decisions are determined by brain activity ahead of time.
We found that the outcome of a decision can be encoded in brain activity of prefrontal and parietal cortex up to 10 s before it enters awareness.
This delay presumably reflects the operation of a network of high-level control areas that begin to prepare an upcoming decision long before it enters awareness.

cited by:
Introduction to Neuroeconomics: how the brain makes decisions
Coursera. July 2014

Wilheml Wundt
{Understanding Psychology © 2014. p. 8}


Conscious of the Unconscious
Work with your unconscious, rather than trying to browbeat it into submission.
Jul 30, 2013

Decoding visual object perception from fMRI

fusiform face area (FFA, red) and parahippocampal place area (PPA, blue).
A human observer, who was only given signals from the FFA and PPA of each participant, was able to estimate with 85% accuracy which of the two categories the participants were imagining.

Decoding mental states from brain activity in humans
John-Dylan Haynes and Geraint Rees
Nature Reviews Neuroscience 7, 523-534 (July 2006)

Recent advances in human neuroimaging have shown that it is possible to accurately decode a person’s conscious experience based only on non-invasive measurements of their brain activity. Such ‘brain reading‘ has mostly been studied in the domain of visual perception, where it helps reveal the way in which individual experiences are encoded in the human brain.
The same approach can also be extended to other types of mental state, such as covert attitudes and lie detection. Such applications raise important ethical issues concerning the privacy of personal thought.

cited by:
Introduction to Neuroeconomics: how the brain makes decisions
Coursera. July 2014

Prefrontal lobe regions & decision making

Prefrontal lobe regions of human brain activated during decision making.

Brains and Decision Making
in: Principles of Animal Communication
Bradbury & Vehrencamp
© 2011 Sinauer Associates, Inc.

figure used by:
Introduction to Neuroeconomics: how the brain makes decisions
Coursera. July 2014

My Stroke of Insight

A Brain Scientist With A ‘Stroke Of Insight’
May 14, 2009

Jill Bolte Taylor was in her late 30s when a blood vessel exploded in her brain.

The irony? Taylor is a neurological researcher.

While a stroke typically leaves devastating effects in the body — and oftentimes leads to death — Taylor has made a complete recovery. She says the experience provided unexpected wisdom.

Her bestselling memoir My Stroke of Insight: A Brain Scientist’s Personal Journey will be published in paperback this month.

… to these individuals, you know, I come right back to the – I think most important message of my whole journey is our human brain is resilient. It is designed to heal itself. I firmly believe that.

And you can try to re-teach new cells in order to feel that again and in order to create new function where you have had cells that have been lost.

cited by:
Coursera Understanding the Brain:

Representation of Scene Categories in the Visual Cortex

Natural Scene Statistics Account for the Representation of Scene Categories in Human Visual Cortex.
Neuron, Volume 79, Issue 5, 1025-1034, 08 August 2013
Stansbury DE, Naselaris T, Gallant JL.


  • Brain activity encodes scene categories that reflect real-world object statistics
  • These scene categories are represented in many anterior visual regions of interest
  • Scene categories and individual objects can be decoded from measured brain activity
  • fMRI signals contain more information about categories than previously appreciated

During natural vision, humans categorize the scenes they encounter: an office, the beach, and so on.
These categories are informed by knowledge of the way that objects co-occur in natural scenes.
How does the human brain aggregate information about objects to represent scene categories?
To explore this issue, we used statistical learning methods to learn categories that objectively capture the co-occurrence statistics of objects in a large collection of natural scenes.
Using the learned categories, we modeled fMRI brain signals evoked in human subjects when viewing images of scenes.

We find that evoked activity across much of anterior visual cortex is explained by the learned categories.
Furthermore, a decoder based on these scene categories accurately predicts the categories and objects comprising novel scenes from brain activity evoked by those scenes.
These results suggest that the human brain represents scene categories that capture the co-occurrence statistics of objects in the world.