Algorithms of the Mind
medium.com
Algorithms of the Mind
1–10 of 20 posts
Re: Algorithms of the Mind
#2Interesting overview of the recognition/imagination duality, but I dislike the tendency to play the game of "oh that's totally what [famous person] must have meant with his dense prose hundreds of years ago."
Re: Algorithms of the Mind
#3The field of computational cognitive neuroscience studies the mind as a machine learning algorithm.
Re: Algorithms of the Mind
#4There are two contradictory claims:
1. The brain is like a neural network (which is purely logical) in the sense of ML.
2. Human brains cannot be explained by purely logical things.
The author also uses "concept," which is a technical term in computational learning theory with a specific meaning, as if it meant "intuition." How can you present "intuition" to a neural network? This distinction is swept under the rug. Not to mention all the recent work showing how easily neural networks can be fooled by slightly adversarially noisy inputs.
There are many grains of salt required for a useful discussion on neural networks. Instead of taking something we have no understanding of and making grand philosophical claims, we should be using the tools we have to understand that thing.
Re: Algorithms of the Mind
#5I think this is an incredibly bad way to try and study the mind. The neural net bears some passing resemblance to a neuron (both have graph connectivity), but the neuron is a biological structure with complex biochemical inputs and outputs. In addition it took us twenty or so years to proceed from simple feed forward neural networks to so-called "deep learning" neural networks. How shallow such networks are when measured against the complexity of an actual neural system is unknown. We may be standing at the shore of a great ocean with one foot in the water congratulating ourselves on our understanding.
Re: Algorithms of the Mind
#6While intriguing, it's important to remember that humankind has always compared the mind to whichever recent technology was available - the catapult, the mill, the steam engine, and eventually, computers.
While Deep Neural Networks -- unlike mills -- are of course inspired by what seems to be the actual biology of our brains, and the results are fascinating, it's humbling to keep the above in mind.
Re: Algorithms of the Mind
#7There are two contradictory claims: 1. The brain is like a neural network (which is purely logical) in the sense of ML. 2. Human brains cannot be explained by purely logical things. The author also uses "concept," which is a technical term in computational learning theory with a specific meaning, as if it meant "intuition." How can you present "intuition" to a neural network? This distinction is swept under the rug.…
Very much agreed. For one thing, comparing the human brain to a deep neural network leaves out the fact that the human brain mostly performs unsupervised perceptive learning, unsupervised causal induction, and reinforcement learning. None of these resemble the deep backpropagation done in most ML models.
Re: Algorithms of the Mind
#8While intriguing, it's important to remember that humankind has always compared the mind to whichever recent technology was available - the catapult, the mill, the steam engine, and eventually, computers. While Deep Neural Networks -- unlike mills -- are of course inspired by what seems to be the actual biology of our brains, and the results are fascinating, it's humbling to keep the above in mind.
I can see how one might talk in parallels between the mind and a mill, or steam engine, or a computer. I don't see how it would work with a catapult, even in a historical context. Can you elaborate? Or even better, if you could show a reference to that.
Re: Algorithms of the Mind
#9I was immediately put off when the author trotted out Sapir-Whorf; and, not even apologetically: in its strong form! Everything in the article became suspect. S-W is not correct, end of story.
Re: Algorithms of the Mind
#10The author of this article fails to incorporate two relevant prior explorations of this topic: (1) from the Buddhist perspective and (2) from Wilfrid Sellars' work, in particular "Empiricism And The Philosophy Of Mind". The remarks below pertain to the first; the second is beyond my philosophy-fu to say anything meaningful.
Take the idea, "We see with our brains, not with our eyes" as a criticism of the "naive view" that sense data / fabrications are neutral, that they are just "out there", and it is only when they come into contact with the mind that the mind infuses the raw sense data with desire and aversion. The idea that we are just passive observers of phenomena.
Thanissaro Bhikkhu critiques this idea from the Buddhist perspective:
"040920 Disenchantment & Dispassion \ \ Thanissaro Bhikkhu \ \ Dhamma Talks" https://www.youtube.com/watch?v=k8M-_Msav1Q
He says that on the contrary, desire and aversion are involved a priori in the formation of the fabrications (sense data).
So this is not a new idea. It is a very old idea. The idea that the technology of ML can confirm this particular critique of the naive view is novel (although I'm not convinced it is wise to draw conclusions about the mind in this way, just as I'm not convinced it is wise to draw conclusions about the way evolution operates based on artificial life simulations).