Author
39 posts
Posts by David Rawlinson.
In the last blog we looked at the phenomenon of “jagged intelligence” — the paradox that AIs are simultaneously superhuman and stupid in adjacent skills. We identified a lack of continual, fast learning and embodiment as possible causes of this weirdness, and noted that embodiment is also a “learning problem”, because current ML models can’t learn from a single thread of…
Introduction Will highly intelligent machines inevitably think like us? Our experiences with AI that is simultaneously incredibly powerful and incredibly flawed are evidence that not all forms of intelligence are the same. Until recently, all human-level intelligence was human. Now, AI seems to threaten our monopoly. But truly human-level and human-like artificial general intelligence (AGI) is still quite…
Authors: Hoang Long Dang and David RawlinsonUPDATE: Full HRM-Agent paper now available. When the Hierarchical Reasoning Model was released, we were very excited about its combination of small size and impressive reasoning abilities. These qualities seem to be enhanced by recurrent processing, but with a surprisingly simple gradient update based on the converged recurrent forward state — turning a…
Since the arrival of Chat GPT, I’ve felt that progress in AI has been largely incremental. Of course, even Chat GPT was/is mostly just a scaled up Transformer model, which was introduced by Vaswani et al in 2017. The belief that AI is advancing relatively slowly is a radical and unpopular opinion these days, and it’s been…
Check out this fascinating research meeting with Numenta and guest speaker Max Bennett. Max describes this thalamocortical microcircuit theory. More detail can be found in his paper, published here: https://www.frontiersin.org/articles/10.3389/fncir.2020.00040/full#h1
We’ve just undertaken a review and refresh of our research roadmap! The topics and approach we will take in the coming year are all laid out in a new page: Research Roadmap Our primary topics for the coming year include: Continual Few-Shot Learning (CFSL) via our Episodic memory system Using reinforcement learning to control a…
Research Roadmap The diagram above shows our research roadmap for the coming year (July 2020-June 2021). You can compare to last year’s diagram (below) to see our progress. Red filling indicates completion; yellow indicates areas we expect to make progress this year. The map has two principal stages – improved representations, and generating behaviour using…
The Whole Brain Architecture Initiative (WBAI) aims to foster research into architectural approaches to general intelligence. They have held a series of events – the Hackathons – that encourage and support the development of new models which provide working solutions to questions about the possible architecture of general intelligence in the brain. You can read…
Artificial neural networks (ANNs) – are conceptually simple; the combination of inputs and weights in a classical ANN can be represented as a single matrix product operation followed by an elementwise nonlinearity. However, as the number of learned parameters increases, it becomes very difficult to train these networks effectively. Most of the tricks that enable…
One of our key projects is a memory system that can learn to associate distant cause & effect while only using local, immediate & unsupervised credit assignment. Our approach is called RSM – Recurrent Sparse Memory. We recently uploaded a preprint describing RSM. This is the first of several blog posts about the experiments we…
We’ve uploaded a new paper to arXiv presenting our algorithm for biologically-plausible learning of distant cause & effect using only local and immediate credit assignment. This is a big step for us – it ticks almost all our requirements for a general purpose representation. The training regime is unsupervised & continuous. Local and immediate credit…
As AI/ML researchers, we have obviously pondered the risks of AI. We even wrote about it. But what might surprise you is the risks that keep AI/ML folk awake at night aren’t the ones you hear about in the media. We’re not worried about runaway “paperclip maximizers” or “skynet”-style machine coups. Andrew Ng says: “worrying…
We recently talked about Capsules networks and equivariances. NB: If you’re not familiar with Capsules networks, read this first. Our primary objective with Capsules networks is to exploit their enhanced generalization abilities. However, what we’ve found instead raises new questions about how generalization can be measured and whether Capsules networks are actually good at it. PredCaps…
We are exploring the nature of equivariance, a concept that is now closely associated with the capsules network architecture (see key papers Sabour et al, and Hinton et al). Machine learning representations that capture equivariance must learn the way that patterns in the input vary together, in addition to statistical clusters in the input (that a typical…
This is the second part of our comparison between convolutional competitive learning and convolutional or fully-connected sparse autoencoders. To understand our motivation for this comparison, have a look at the first article. We decided to compare two specific algorithms that tick most of the features we require: K-Sparse autoencoders, and Growing-Neural-Gas-with-Utility (GNG-U) (Fritzke1997). The latter…
Competitive learning is a branch of unsupervised learning that was popular a long, long time ago in the 1990s. Older readers may remember – the days before widespread use of GSM mobile phones and before Google won the search engine wars! Although competitive learning is now rarely used, it is worth understanding the broader range of…
We’ve just uploaded a spin-off research paper to arXiv titled “Sparse Unsupervised Capsules Generalize Better”. So what’s it all about? Capsules Networks You may have heard of Capsules Networks already – if not, have a read of one of these blog articles (here, here, here, or here (EM routing)), watch this video, or consult one of the…
We attended the 2017 10th Conference on Artificial General Intelligence, which was located in our hometown of Melbourne, Australia! Excitingly, the IJCAI 2017 conference is also in Melbourne this week and ICML 2017 was in Sydney this year. In particular, the “Architectures for Generality and Autonomy” workshop may be of interest to readers. You can find…
Today’s post tries to fit the theoretical concept of Predictive Coding with the unusual structure and connectivity of Pyramidal cells in the Neocortex. Pyramidal neurons Pyramidal neurons are interesting because they are one of the most common neuron types in the computational layers of the neocortex. This almost certainly means they are critical to many…
Typical results from our experiments: Some active cells in layer 3 of a 3 layer network, transformed back into the input pixels they represent. The red pixels are positive weights and the blue pixels are negative weights; absence of colour indicates neutral weighting (ambiguity). The white pixels are the input stimulus that produced the selected…
Figure 1: The Region-Layer component. The upper surface in the figure is the Region-Layer, which consists of Cells (small rectangles) grouped into Columns. Within each Column, only a few cells are active at any time. The output of the Region-Layer is the activity of the Cells. Columns in the Region-Layer have similar – overlapping –…
What’s AlphaGo? AlphaGo is a system that can play Go at least as well as the best humans. Go was widely cited as the hardest (and only remaining) game at which humans could beat machines, so this is a big deal. AlphaGo has just defeated a top-ranked human expert. AlphaGo Nature paper (Silver et al 2016)…
The artificial neuron model used by Jeff Hawkins and Subutai Ahmad in their new paper (image reproduced from their paper, and cropped). Their neuron model is inspired by the pyramidal cells found in neocortex layers 2/3 and 5. It has been several years since Jeff Hawkins and Numenta published the Cortical Learning Algorithm (CLA) whitepaper.…
Authors: D Rawlinson and G Kowadlo This is the first of three articles detailing our latest thinking on general intelligence: A one-size-fits-all algorithm that, like people, is able to learn how to function effectively in almost any environment. This differs from most Artificial Intelligence (AI), which is designed by people for a specific purpose. This…
An interesting article by Gerard Rinkus comparing the qualities of sparse distributed representation and quantum computing. In effect, he argues that because distributed representations can simultaneously represent multiple states, you get the same effect as a quantum superposition. The article was originally titled “sparse distributed coding via quantum computing” but I think that gets the…
By David Rawlinson and Gideon Kowadlo The aim of this post is to take a short departure from more technical issues and deal with some of the philosophical questions about Artificial General Intelligence (AGI). The Elephant in the room This metaphor implies willful ignorance of something blindingly obvious. Something that’s difficult to ignore. It requires effort…
by David Rawlinson and Gideon Kowadlo Why the retina is worth a deeper look Recently, we have been looking at the retina – light sensitive cell layers inside the eyeball, that detect wavelength and intensity, and compute some initial encoding of this data that is transmitted to the brain proper via the optic nerve. The…
This might be our new test suite for the algorithms! http://www.arcadelearningenvironment.org/ “The Arcade Learning Environment (ALE) is a simple object-oriented framework that allows researchers and hobbyists to develop AI agents for Atari 2600 games. It is built on top of the Atari 2600 emulator Stella and separates the details of emulation from agent design.” Why…
If you’re looking for a formal and precise description of the current HTM/CLA algorithm, then this blog [1, 2] by Fergal Byrne may be of interest. Our experience is that while pseudocode is a good way of describing an algorithm, equations can often be more succinct and specific. The CLA whitepaper [3] describes the algorithm…
This post asks some questions about the agency of hierarchical action selection. We assume various pieces of HTM / MPF canon, such as a cortical hierarchy. Agency The concept of agency has various meanings in psychology, neuroscience, artificial intelligence and philosophy. The common element is having control over a system, with varying qualifiers regarding the entities…
This post is about a fantastic new paper by Michael R. Ferrier, titled: Toward a Universal Cortical Algorithm: Examining Hierarchical Temporal Memory in Light of Frontal Cortical Function The paper was posted to the NUPIC mailing list and can be found via: http://numenta.org/community-content.html The paper itself is currently hosted at:…
This is a quick post to link a poster paper by Ryan McCall, who has experimented with a Predictive-Coding / Cortical Learning Algorithm (PC-CLA) hybrid approach. We found the paper via Ryan writing to the NUPIC theory mailing list. What’s great about the paper is it links to some of the PC papers we mentioned…
Browsing the NUPIC Theory mailing list, I came across a post by Fergal Byrne on the differences and similarities between Deep Learning and MPF/HTM. It’s a great background into some of the pros and cons of each. Given the popularity and demonstrated success of Deep Learning methods it’s good to understand how they work and how…
Introduction Predictive Coding (PC) is a popular theory of cortical function within the neuroscience community. There is considerable biological evidence to support the essential concepts (see e.g. “Canonical microcircuits for predictive coding” by Bastos et al). PC describes a method of encoding messages passed between processing units. Specifically, PC states that messages encode prediction failures; when…
A quick post here to give readers an idea of HTM/CLA performance on visual object recognition tasks. Although a little out of date now, Craig’s blog gives a good idea of what NUPIC can and can’t do, and how data issues (e.g. training set, variations in object appearance) affect the results. This is a handy…
by David Rawlinson and Gideon Kowadlo A Temporal Pooler ProposalWe will describe a Temporal Pooler design below. We have tested this design and it gives satisfactory output given the criteria and expectations described below and in the previous articles in this series (post #1, post #2). It is also relatively simple. You could imagine this…
By David Rawlinson and Gideon Kowadlo Jeff’s new Temporal Pooler This is article 2 in a 3 part series about Temporal Pooling (TP) in MPF/CLA-like algorithms. You can read part 1 here. For the rest of this article we will assume you’ve read part 1. This article is about the new TP proposed by Jeff Hawkins. The…
by David Rawlinson and Gideon Kowadlo Article Series This is the first of 3 articles about temporal pooler design for Numenta’s Cortical Learning Algorithm (CLA) and related methods (e.g. HTM/MPF). It has limited relevance to Deep Belief Networks. This first part will describe some of the considerations, objectives and constraints on temporal pooler design. We…
by David Rawlinson and Gideon Kowadlo Introduction The Memory Prediction Framework (MPF) is a general description of a class of algorithms. Numenta’s Cortical Learning Algorithm (CLA) is a specific instance of the framework. Numenta’s Hierarchical Temporal Memory (HTM) was an earlier instance of the framework. HTM and CLA adopt different internal representations so it is…