[HN Gopher] Deep learning meets vector-symbolic AI
___________________________________________________________________
Deep learning meets vector-symbolic AI
Author : sayonaraman
Score : 59 points
Date : 2021-09-23 16:24 UTC (6 hours ago)
(HTM) web link (research.ibm.com)
(TXT) w3m dump (research.ibm.com)
| uoaei wrote:
| Seems like a variant of a Siamese network which uses binarized
| embedding vectors for predictions instead of the raw embedding
| vectors. What exactly is the novelty presented here?
| axiosgunnar wrote:
| Can any experts chime in? Is this any good?
| armcat wrote:
| The fundamental idea is "operating in high dimensions", and
| this does have some solid footing, e.g. see Cover's Theorem
| (https://en.wikipedia.org/wiki/Cover's_theorem). In fact I
| recently did a presentation of another paper (from FB AI
| Research and Carnegie Mellon), exploring the concept of
| projections to higher dimensions for sentence encoding tasks,
| see here: https://github.com/acatovic/paper-
| lunches/blob/main/fb-rands....
|
| There is fair amount of research in the area of high
| dimensional computing, as well as with sparse representations,
| which seem to be grounded in neuroscience. As others have
| pointed out, a number of commercial research labs exists. There
| is Numenta with their Hierarchical Temporal Memory, and
| Vicarious (whose founder was one of Numenta's co-founders), as
| well as Cortical.io (who are borrowing sparse binary coding
| concepts from Numenta, combining it with self-organizing maps,
| and applying it to document understanding tasks).
| sayonaraman wrote:
| this is awesome, thanks for the link! There seems to be
| mutual propagation from NLP to CV and from CV to NLP, i'm
| wondering if there is a visual counterpart for these "Random
| Encoders". The SOTA for image/text embeddings currently seems
| to be CLIP [1].
|
| [1] https://openai.com/blog/clip/
| robbedpeter wrote:
| Different words, same concept as sparse distributed
| representation in a hierarchical distribution of neural
| networks.
|
| Automating the process of learning is non-trivial and making it
| efficient is an ongoing question.
|
| Vicarious.ai , Hawkins Numenta, cortical.io and various other
| projects have been chasing this in various guises for many
| years.
|
| The lottery ticket effect on very large networks can make
| contrasting different architectures difficult, and this looks
| suspect. IBM isn't necessarily a powerhouse in this arena, so
| it'd make sense not to get excited until they verify and expand
| their theory. It could be that their initial success is
| entirely coincidental with lotteries and the particulars of the
| design are a dead end.
| sayonaraman wrote:
| you might also be interested in the recent work on "resonator
| networks" VSA architecture [1-4] by Olshausen lab at Berkeley (P.
| Kanerva who created the influential SDM model [5] is one of the
| lab members).
|
| It's a continuation of Plate [6] and Kanerva work in the 90s and
| Olshausen' groundbreaking work on sparse coding [7] which
| inspired the popular autoencoders [8].
|
| I find it especially promising they found this superposition
| based approach to be competitive with optimization so prevalent
| in modern neural nets. May be backprop will die one day and be
| replaced with something more energy efficient along these lines.
|
| [1] https://redwood.berkeley.edu/wp-
| content/uploads/2020/11/frad...
|
| [2] https://redwood.berkeley.edu/wp-
| content/uploads/2020/11/kent...
|
| [3] https://arxiv.org/abs/2009.06734
|
| [4] https://github.com/spencerkent/resonator-networks
|
| [5] https://en.wikipedia.org/wiki/Sparse_distributed_memory
|
| [6] https://www.amazon.com/Holographic-Reduced-Representation-
| Di...
|
| [7] http://www.scholarpedia.org/article/Sparse_coding
|
| [8] https://web.stanford.edu/class/cs294a/sparseAutoencoder.pdf
___________________________________________________________________
(page generated 2021-09-23 23:01 UTC)