Wednesday, April 17, 2013

The Perfect Brain: Another Nail in the Coffin of the Bayesian Brain

The Impending Crash of the Bayesian Bandwagon

Last August, I wrote a series of posts titled, The Myth of the Bayesian Brain. I argued against the prevailing notion in the AI community that the brain uses some kind of Bayesian statistics to make decisions. I argued that, internally, the brain always assumes that the world is perfect even if its sensory space is inherently noisy. The brain does this bit of magic by filling in any missing information and ignoring irrelevant noise. This cleansing process is essential to reasoning and planning. At least one other researcher (to my knowledge), computer scientist Judea Pearl, has been saying the same thing. Well, a story out of Princeton University points to a new study that corroborates what I have been saying. Essentially, Princeton University researchers found that, when we make an error, the brain's decision making system is not at fault. The system is flawless. The fault is invariably due to faulty sensory information. Here's an excerpt:
Previous measurements of brain neurons have indicated that brain functions are inherently noisy. The Princeton research, however, separated sensory inputs from the internal mental process to show that the former can be noisy while the latter is remarkably reliable, said senior investigator Carlos Brody, a Princeton associate professor of molecular biology and the Princeton Neuroscience Institute (PNI), and a Howard Hughes Medical Institute Investigator.

"To our great surprise, the internal mental process was perfectly noiseless. All of the imperfections came from noise in the sensory processes," Brody said.
The "great surprise" of Carlos Brody and his team is understandable, given their training within the current Bayesian paradigm. But it's never too late to jump off that silly wagon. I am not one to laugh and say, "I told you so". But I did, didn't I?

See Also:

The Second Great AI Red Herring Chase
The Myth of the Bayesian Brain


Bill said...


Arcadia said...

What'd you think about this:

Louis Savain said...

Arcadia wrote:

What'd you think about this:

I think it's crap. Sorry to be so blunt. Any time you see people invoke mathematics as the way to solve intelligence, you can be absolutely certain that it's crap. The whole Singularity approach to AI is steeped in math. It's all crap, in my opinion. Again, I apologize for being so dismissive in my criticism. It's my way of focusing on what I think is a better approach.

PS. I must add that, if you are a Singularitarian, I retract my apology.

Arcadia said...

Not sure why you concluded it is singularity. The whole idea is physics and entropy/thermodynamic where intelligence emerged, it's god's work.

anyway, forbes has written a complete story about it,

Louis Savain said...


Hi. The reddit post mentioned Singularity and how the work of Wissner-Gross resembles what Jason Silva has been talking about on YouTube. I now realize that Wissner-Gross is not associated with the Singularity movement. No matter. Here's an excerpt from the Inside Science article:

"[The paper] is basically an attempt to describe intelligence as a fundamentally thermodynamic process," said Wissner-Gross.

Huh? In my opinion, this stuff is obviously a con game. I was not 100% sure at first but then I found this little gem near the end of the article:

Wissner-Gross suggested that the new findings fit well within an argument linking the origin of intelligence to natural selection and Darwinian evolution -- that nothing besides the laws of nature are needed to explain intelligence.

That's what did for me. At that point, there was no doubt in my mind that it is pure unmitigated bovine excrement. Sorry. Those guys should be prosecuted for fraud or something. I left a piece of my mind in the comments section of the Forbes article. He's welcome to sue me if he wants.

Louis Savain said...

Just in case some of you are wondering why I think this entropic theory of intelligence is crap, here's a brief refutation:

Wissner-Gross claims that intelligence is a process that tries to minimize entropy (maximizing order) but the truth is that intelligent organisms don't care about entropy. They only care about maximizing pleasure and minimizing pain regardless of what happens to entropy. If Wissner-Gross was right, wars would not exist.

Another problem is that there is no way one can describe intelligence with a single equation. An intelligent system consists of many interacting modules each with a specific function. For example, we have pattern learning, attentive behavior, sequence learning, invariant object recognition, goal-seeking behavior, adaptation, etc. Moreover, these things require a hierarchical memory to work properly, and this has nothing to do with entropy.