From Deep Learning of Disentangled Representations to Higher-level Cognition

One of the main challenges for AI remains unsupervised learning, at which humans are much better than machines, and which we link to another challenge: bringing deep learning to higher-level cognition. We review earlier work on the notion of learning disentangled representations and deep generative models and propose research directions towards learning of high-level abstractions. This follows the ambitious objective of disentangling the underlying causal factors explaining the observed data. We argue that in order to efficiently capture these, a learning agent can acquire information by acting in the world, moving our research from traditional deep generative models of given datasets to that of autonomous learning or unsupervised reinforcement learning. We propose two priors which could be used by an agent acting in its environment in order to help discover such high-level disentangled representations of abstract concepts. The first one is based on the discovery of independently controllable factors, i.e., in jointly learning policies and representations, such that each of these policies can independently control one aspect of the world (a factor of interest) computed by the representation while keeping the other uncontrolled aspects mostly untouched. This idea naturally brings fore the notions of objects (which are controllable), agents (which control objects) and self. The second prior is called the consciousness prior and is based on the hypothesis that our conscious thoughts are low-dimensional objects with a strong predictive or explanatory power (or are very useful for planning). A conscious thought thus selects a few abstract factors (using the attention mechanism which brings these variables to consciousness) and combines them to make a useful statement or prediction. In addition, the concepts brought to consciousness often correspond to words or short phrases and the thought itself can be transformed (in a lossy way) into a brief linguistic expression, like a sentence. Natural language could thus be used as an additional hint about the abstract representations and disentangled factors which humans have discovered to explain their world. Some conscious thoughts also correspond to the kind of small nugget of knowledge (like a fact or a rule) which have been the main building blocks of classical symbolic AI. This, therefore, raises the interesting possibility of addressing some of the objectives of classical symbolic AI focused on higher-level cognition using the deep learning machinery augmented by the architectural elements necessary to implement conscious thinking about disentangled causal factors.
See more at www.microsoft.com/en-us/resea...

[@johntanchongmin]

Interpolating in abstract space, exactly what Stable Diffusion is doing. This idea is really impactful.

[@itaybenou]

Wonderful video. You can't help but admire his approach for what is AI, and the way he manages to convey these concepts. Brilliant!

[@flamingxombie]

The intuition for why we current speech models can't produce good unconditional samples (see wavenet) is simply mindblowing. Phonemes occupy a small number of bits as compared with the overall signal (~10/s as compared with 16 k/s)!

[@mjParetoQuant]

I liked that point very much. However, it is also quite obvious: we don't analyse the signal with our brain, we analyse the sentenses and the meaning.

[@bingeltube]

Thanks for the interesting talk! Please post the slides as well!

[@ProfessionalTycoons]

amazing talk.

[@silberlinie]

Deepening of learning into a higher cognitive level: Very good. What and where are the works, who is working on this approach?

[@catsaresocute650]

In Lexes Podcast just the mention of the example of speech as becoming unrecognicable gibberish (due to the amount of data) but when you seperate the gibberish to get a baisc feel for intonation or sound and speach as vocilisation of cerain tones that humans think of as speach you get a functional way to work it out would have totally suffieced to get the thing

[@tempvariable]

In 12:07, are cognitive states low dimensional if that is the case are they sparse? If they are both sparse and low dimensional it contradicts with what he said in his MSS talk in 2012, where he states high dimensional and sparse is better than low dimensional

[@maloxi1472]

He's allowed to change his opinion and improve his theories over time. That's (interestingly enough) the kind of stuff that general intelligence allows

[@rahuldeora5815]

Someone should write a detailed blog explaining stuff in this

[@siarez]

Who is the gentleman at 1:09:35 asking a question, and bringing up gradual learning?

[@DeadRabbittt]

Its Patrice Simard from Microsoft Research

[@nguyenngocly1484]

You can turn artificial neural networks inside-out by using fixed dot products (weighted sums) and adjustable (parametric) activation functions. The fixed dot products can be computed very quickly using fast transforms like the FFT. Also the number of overall parameters required is vastly reduced. The dot products of the transform act as statistical summary measures. Ensuring good behavour. See Fast Transform (fixed filter bank) neural networks. The variance equation for linear combinations of random variables is very useful for understanding dot products in neural networks especially in conjunction with cosine angle. Also ReLU is a switch. The electricty in your house is a sine wave. Turn on a switch and the output is f(x)=x. Again the same sine wave as the input. Off(x)=0. A ReLU neural network then is a switched composition of dot products. If the switch states are known then there is a linear mapping between the input vector and the output vector which you can check out with various metrics.

[@ahmadchamseddine6891]

He is genius

[@johntanchongmin]

In 1:09:33, there was a question on gradual change in hypothesis space from very few samples - theory revision. I feel like neural nets may be quite ill-suited for fast change of learnt knowledge as the weights take a long time to change by backpropagation. What I believe is necessary, will be to imbue some form of learnable external memory bank on which we draw our knowledge from (in addition to neural nets), so we can just change that knowledge bank and learn new concepts instantly.

[@juliocardenas4485]

The camera work negatively affects a wonderful lecture

[@muckvix]

Anyone has a link to the slides? And come on camera people, it's not a beauty pageant, it's ok if you show slides instead of the speaker's face :)

[@oudarjyasensarma4199]

@@ProfessionalTycoons 404 not found!!! Can you share a revised link? Thanks in advance!!!

[@reidalbecker3830]

medium.com/@SeoJaeDuk/archived-post-from-deep-learning-of-disentangled-representations-to-higher-level-cognition-b848fdc0de2c

[@ewfq2]

I want to talk with the guy talking about barycentres and wasserstein distance!

[@ewfq2]

around 1:06:00

[@ewfq2]

What's the neurips optimal transport tutorial mentioned?

[@scose]

Sampling rate * bit depth is a big overestimate of the amount of information in speech audio signals - look at the compression ratios that audio codecs can achieve

[@jonabirdd]

51:00 I like the idea of a two-level system but disagree with the mutual information criterion.

[@nathanbittner1452]

Interesting. Could you say a bit more on this?

[@dr.mikeybee]

Doesn't translation into an abstract space necessitate a loss of information?

[@tomm7273]

Yes, but the benefits of dimensionality reduction far outweigh that. You don't need to consider every pixel of a picture to reason about the objects contained within that picture and their features.

[@zachundisclosed6706]

There is also information stored in the decoder.

[@nauy]

No, not if the information content is low dimensional to begin with. Consider a circle of radius r is rendered at location (x, y) on a bit map. The information in pixel space is high dimensional - the number of pixels in the bit map. But the same circle can be transformed into a 3-dimensional parameter space representation - (x, y , r) with no loss of information. The same circle in pixel space can always be regenerated using the parameter space representation.

[@dr.mikeybee]

@@nauy Thanks. I've been learning about matrix transformations and PCA lately. It took me a few years to get here.

[@runvnc208]

Sounds right to me. But why do they assume that the traditional neural net and deep learning are the best or only possible fundamental structures and processes for a system with these capabilities of disentangled abstractions working together with granular representations?

[@moisesfelipe9596]

Good point. I'd like to hear something in that sense. I fell the current popularization of NN and DL has lead a lot of people to not consider any other alternatives and then losing other ways to solve problems and useful insights.

[@moisesfelipe9596]

For example, when I heard the term disentangling I cannot stop thinking on this as a fancy (and potentially more sophisticated) way to refer to blind source separation.

[@MartinLichtblau]

Humans use fuzzy approaches, while computers use precise numbers. Which one can work in this complex world?

[@tomm7273]

Computers can use fuzzy approaches as well. Almost all modern machine learning techniques are fuzzy.

[@MartinLichtblau]

​@@tomm7273 If you mean Deep Learning I'd say: yeah the direction seems ok. But the way computers work, and any representation they use is, quantitive and they are absolutely precise with those numbers. While humans think in qualitative terms, like "this rough concept is very similar to that one". Indeed they can't quantify things precisely, but that is what makes humans more capable to deal with all this ambiguous complexity.

[@ahilanpalarajah3159]

@@MartinLichtblau Why do you think fuzziness has to contradict precise numbers? I'm not arguing it doesn't, I'm just asking because we can fuzzify and work with vagueness to eliminate as much of the search space as possible.

[@MartinLichtblau]

@@ahilanpalarajah3159 It's complicated, but in it's basic sense it doesn't. I just couldn't find simple terms to tell them apart. Perhpas better say Accurate vs. Approximate or rigid vs. flexible...

[@mikepict9011]

Humans have emotion , humans care .... robots will never care

[@yeodongyoun6780]

Wow the subtitles are terrible :( ... GAN -> gown, k means -> keys lol

[@zlh]

46:00 re: attention as gating the conscious and unconscious thoughts - can you imagine a machine which can widen and narrow its aperture of attention to accomplish different tasks?

[@rahuldeora5815]

Someone should write a detailed blog explaining stuff in this