FIRST and foremost thank you for putting this forum together. I have been waiting this format forever!
Anyway my name is Mohamed Bennouf and live in the Bay Area (US) I was raised in France but been in the States since 1988.
I have a passion for AI since I was little kid. 2001 is of course my favorite movie. I then found Jeff’s book and all I could say at the time was WOW! It made some much sense. Eventually Numenta came up with a very easy software (HTM 1.0) and the amazing web services which allowed to experiment with different problems. Loving mobile apps, I was able to for instance combine the two and come up with an app that can learn to recognize couple type of birds and say the name of the bird out loud. That was so much to do Here an old video of the system I made:
Recently Numenta came up with a much more powerful algo (CLA) and i have to admit I did not spend much time on it but I can see how powerful it is versus HTM 1.0 The problem is that I was never able to install the software on my Macbook! Even so there is an amazing tutorial video from Numenta. I just cannot understand all the install steps. Time is also a big issue of course. I will LOVE to finally install Nupic and start playing with it like I did with HTM 1.0 and the webservices at the time. That is another reason why I am so much excited about this forum. It will be much easier to ask for help
One my interest is to use Nupic for machine self diagnostic. One scene that is printed in my brain in the 2001 Space Odyssey was when Hal 9001 told the astronauts that the main antenna on the spaceship was going to fail in 72 hours! I truly believe Nupic can achieve that. A system like that could save a huge amount of money to companies. I am field service engineer (mass spectrometer) so I would love Nupic to help me…
That’s it. Can wait to converse with all the smart people in this forum>
Have a great week all>
Not to discourage you from installing the Python version (if that’s your thing), but if you like, you can run the Java version (htm.java) which requires no installation? It is the same as NuPIC though it is playing “catch up” since it is primarily community supported.
See the examples and demos here
Matt (@rhyolight) can help you get the Python version running too - especially for MacBooks it should be much simpler.
My name is Sam Gallagher, and I’m an electrical engineering student at Drexel University. I read On Intelligence when I was in high school a good few years ago, and I really liked it! It has been a favorite book of mine, and I have kept up-to-date to the best of my ability over these last five years or so paying attention to the development of the CLA and nupic. I am not a well-educated PhD or neuroscientist, but I’m a passionate engineer and aspiring inventor. What brought me to really concentrate on HTM is the initial idea of self-driving cars, which I had been planning to work on a couple years ago. I was in high school though, blithely unaware of how cars worked! So I missed that boat somewhat. But here I am, working with HTM even still. I have been writing my own CLA in MATLAB for the past four months or so, though I am not a computer scientist. I still could not be happier with how the program works (almost done with temporal memory, had many issues!) and I hope to be able to put some fine touches on it with the help of this forum.
I can’t wait to work with everyone to learn NuPIC and develop the theory of HTM further!
My name is Hanbaek, a Ph.D candidate in math at the Ohio State University.
I study distributed algorithms and complex dynamical systems, mainly discrete ones, as known as cellular automata.
I has always been excited about complex systems that can exhibit surprising high-level functions through some non-linear sum of simple local interactions; bird and fish flocking, unison of firefly blinking, and the brain!
When I read On intelligence in my junior year, it totally got me. The Memory-prediction scheme was so elegant a principle for how the brain works. Now we have deeper understanding of how to construct such a pattern-learning complex system; use of SDRs, Encoders, and temporal memory etc., which is really exciting.
As Von Neumann invented cellular automata as a simplified mathematical model for self-replicating organisms, my hope is to construct a (hopefully simple) mathematical system that can learn and predict sequential data and mathematically prove that it really does. This is a long goal and till then I will have to learn a lot from this awesome community and think together with you guys.
This is a FANTASTIC news! I downloaded the files in:
I can see the Hotgym example for instance but how to run it? I have Xcode install (i do ios apps) so the code opens automatically using xcode but not sure how run it? Do I need to install something else like Eclipse for instance?
In any event, if I do not need to mess with the Pyhton version that would be fantastic
Thanks again David.
I have opened a new topic for this, as we should keep this thread on course to be about “introductions”
So sorry Yes that makes a lot of sense, thank you!
Now I want to watch 2001 A Space Odyssey!
I currently have an idea for an AI search engine for the internet that uses HTM memory. It would have image recognition, language recognition, and user profiling. It would take a search text and try to find what you’re looking for based on what you’ve liked before. I don’t know if this program will ever get off the ground but I’m hoping for it. It would use language processing to “read” text on the web and then profile it using its model of what I like. I’ve never written a search engine before so this is fun. I’m a new programmer who’s just learning but I figure it’s never too late to start!
Hi, I am cofounder and CTO of an AI consulting company. We specialise in the design and deployment of deep learning and cognitive frameworks for enterprises - www.dsp.ai. A long time ago now, I was enrolled in the PhD in physics program at the University of Massachusetts at Amherst, studying particle physics. Since then I have been a Systems Engineer at startups and MNC’s, and continued my education in business (I have an MBA), financial trading algorithms, and most recenty in AI. I see the development and application of AI in business and embedded in society keeping us busy for the foreseeable future .
Have been interested in the human brain and machine intelligence for decades.
As undergrad worked for Christina Enroth mapping retinal receptive fields.
Grad advisor was AI pioneer Donald Michie.
Currently an industrial machine vision engineer, which is unfortunately all about very careful design of lighting, rather than intelligence: http://machinevision.engineering/
Exploring a novel black-box approach to machine learning: generating the most accurate set of conditional PDFs from a sequence of observations of inputs and outputs, and using info theory to guide exploration.
And I may have partially cracked the “hard” consciousness enigma.
Hi, I’m a 5th year software engineering student at the University of Bergen in Norway, working on my master’s thesis. I was interested in machine learning and wished to work on something ML-related on my thesis so I talked with a professor that had knowledge in the field, and it turned out he was very interested and passionate about HTM so he introduced me to it. Once I was hooked on HTM, I decided to work on a thesis that somehow involved using HTM/NuPIC, and it ended up being trying to utilize HTM/NuPIC to forecast the weather (as best as possible).
So here I am today, trying to understand NuPIC to my utmost, and using that knowledge for getting the best possible weather predictions. The way I’m going about this, is by having gathered a lot of historical weather data measurements (daily measurements, but over the course of 10 years) of temperature, precipitation, humidity, air pressure and wind speed. I’m trying to feed each measurement metric into NuPIC (trying both alone and together in correlation with the other measurement metrics) and getting out predictions. So far, my predictions have been so-so, which isn’t very surprising really, since the weather is an extremely complex and vast system dependent on a huge amount of variables and processes way outside the scope of NuPIC’s abilities. I’m gladly open to any constructive suggestions for improving my predictions
Hello everyone, I’m a software engineer at a small dsp company. I have always been interested in the mind and how it learns. Luckily I ran into On Intelligence about 6 months ago. I’ll be slowly going though the tutorials and trying to run my own data as I have a year left in my MSEE in DSP. I look forward to the advancement in this field.
What’s up players?! I’m Roberto. Found On Intelligence around 2004 and it blew my mind. An unsuccessful Masters and a successful sales career later, I’m diving back into the frey. Very excited.
Just as enthusiastic as everyone here, I am in love with AI and the concept of HTM. Hope to have fun along the way with everyone. Cheers!
My research is in the area of Music Information Retrieval and Machine Learning on symbolic music data. I am really interested in the application of Nupic algorithms to music prediction and anomoly detection in music. I also am finishing my PhD in University Technology, Sydney and am have created the interactive music search engine, stelupa.com and I trying to implement some nupic stuff in this application.
If anyone else is into music data data analytics machine learning stuff, would let me know!
4 posts were split to a new topic: HTM for music analysis
My name is Jos Theelen. I built with some of the nupic.core files a 32-bits C+±library, so I could run (parts of) nupic on my old linux-computers. I only use the Spatial Pooler and the Hierarchical Temporal Memory from nupic.core. Other things like encoders I wrote myself.
At jgab3103: One of my HTM-projects is an attempt to let the computer pick chords for a given melody. I took the chorals of the Weinachts-Oratorium from Bach and simplified each of them to a melody with chords. Those note/chords combinations I put in a SP and HTM, so it learned all those combinations. Then I took a new melody and looked which note/chord could be used and which of them had the lowest anomaly value. The results weren’t very good. The chords fitted with the notes of the melody, but the sequence of chords didn’t sound good. My next idea is to program an extra SP and HTM, which uses the output of the first SP and HTM as input. It should somehow learn little pieces of melody, a sequence of note/chord combinations and remember them. With a given melody, I could try to find a sequence of those little pieces of melody.
Got introduced to NUPIC early this summer. It has been a summer of fascinating learning. To learn more deeply about the theory, I decided to implement the algorithm in matlab. I now have a version of the code that produces fairly close anomaly scores to numenta’s version on the NAB benchmark. Some fine tuning still needs to be done.
I benefited greatly from the openness that is present in this community in terms of the willingness to share code, ideas, as well as to document their failures and frustrations.
I would to hear from others who have tried a matlab implementation.
I am a computer vision researcher with more than 25 years of algorithm development. My next goal is to use HTM idea for video analysis.
My name is Rikkert Koppes. Just signed up. I am a software engineer with a background in physics and cognitive ergonomics. I started actively studying everything around AI some 1.5 years ago. Followed the MIT AI course by Patrick Winston and the Oxford Deep Learning course by Nando de Freitas and Hinton’s course on Coursera. Read quite a few books, from the very early ML topics (svm, knn etc) to some more modern views. Around last year I read “how to create a mind” by Kurzweil
Somehow, I completely missed HTM
I have been playing around with languages like Clojure and Prolog, I like how they seem closer to what I feel seems to be the way forward. Some thoughts I have are:
- I feel AI should at least be able to solve the principles from CRUM: rules, logic, concepts, analogies, images, networks (from Thagard, Mind: Introduction to Cognitive Science).
- I feel AI should have some method of introspection. Capture the pattern in a pattern so to say. I don’t see how ideas thus far could do that. What appeals to me in a language like clojure is that the language itself is a data structure. I feel a mind should be similar.
- I like Hinton’s thought vectors, which may not be that dissimilar in concept from SDR’s
- I like the hallucination efforts from DNN’s
- I like how natural HTM theory sounds
I didn’t look too much into the forum yet, but I will. As an HTM newbee, I have a lot to read still. I might play around with some concepts in nodejs (my preferred platform)