# Machine Learning

**URL:** https://discourse.numenta.org/c/engineering/machine-learning/48.md?page=1

[Latest](https://discourse.numenta.org/latest.md) · [Categories](https://discourse.numenta.org/categories.md) · [Tags](https://discourse.numenta.org/tags.md)

**Page:** 2

---

## [Anomaly detection with HTM.core model](https://discourse.numenta.org/t/anomaly-detection-with-htm-core-model/10146)

<div class="topic-metadata">

**Author:** [@Sze\_Lim](https://discourse.numenta.org/u/Sze_Lim)\
**Replies:** 1\
**Last updated:** [October 1, 2022, 10:22pm UTC](https://discourse.numenta.org/t/anomaly-detection-with-htm-core-model/10146 "2022-10-01T22:22:13Z")

</div>

Hi, I am new to HTM and a slow learner, I tried to implement the codes with the HTM.core package with my own dataset and I got this results. It outputs “claLearningPeriod is deprecated, use learningPeriod instead.”, and…

---

## [Have any ML researchers/engineers outside Numenta tried applying its ideas?](https://discourse.numenta.org/t/have-any-ml-researchers-engineers-outside-numenta-tried-applying-its-ideas/9600)

<div class="topic-metadata">

**Author:** [@strangecosmos](https://discourse.numenta.org/u/strangecosmos)\
**Replies:** 5\
**Last updated:** [September 30, 2022, 4:16pm UTC](https://discourse.numenta.org/t/have-any-ml-researchers-engineers-outside-numenta-tried-applying-its-ideas/9600 "2022-09-30T16:16:49Z")

</div>

Have any machine learning researchers or engineers (or other ML practitioners) outside Numenta tried applying any of Numenta’s ideas to machine learning? If not, what would be needed for members of the broader ML commun…

---

## [2D MNIST Encoder](https://discourse.numenta.org/t/2d-mnist-encoder/10073)

<div class="topic-metadata">

**Author:** [@cezar\_t](https://discourse.numenta.org/u/cezar_t)\
**Replies:** 6\
**Last updated:** [September 15, 2022, 1:07pm UTC](https://discourse.numenta.org/t/2d-mnist-encoder/10073 "2022-09-15T13:07:16Z")

</div>

Hi I did some experiments involving SDRClassifier and a few encoders. SpatialPooler, FlyHash Encoder and a newer FH 2D Encoder which combines idea of random projection from fly hash with spatial proximity. Which means …

---

## [Can CNNs serve as weak empirical proof for the thousand brains theory?](https://discourse.numenta.org/t/can-cnns-serve-as-weak-empirical-proof-for-the-thousand-brains-theory/9908)

<div class="topic-metadata">

**Author:** [@Joy\_Lunkad](https://discourse.numenta.org/u/Joy_Lunkad)\
**Replies:** 1\
**Last updated:** [September 8, 2022, 2:24am UTC](https://discourse.numenta.org/t/can-cnns-serve-as-weak-empirical-proof-for-the-thousand-brains-theory/9908 "2022-09-08T02:24:30Z")

</div>

Modelling the thousand brains theory of intelligence In the Article linked below - I presented soft proof of why convolutions can be used as a foundation for cortical mini-columns. I also proposed a novel convolutional …

---

## [Sales Prediction Machine Learning Project](https://discourse.numenta.org/t/sales-prediction-machine-learning-project/9858)

<div class="topic-metadata">

**Author:** [@Arjun\_Kumar](https://discourse.numenta.org/u/Arjun_Kumar)\
**Replies:** 0\
**Last updated:** [June 23, 2022, 6:32am UTC](https://discourse.numenta.org/t/sales-prediction-machine-learning-project/9858 "2022-06-23T06:32:00Z")

</div>

Hello Everyone, I am working on sales prediction project and I am confused about the source code. I have checked on GitHub to find the source code of this project… Basically, this project dataset comprises 2013 sales dat…

---

## [Why does ML use y as input?](https://discourse.numenta.org/t/why-does-ml-use-y-as-input/9659)

<div class="topic-metadata">

**Author:** [@Jose\_Cueto](https://discourse.numenta.org/u/Jose_Cueto)\
**Replies:** 3\
**Last updated:** [April 25, 2022, 5:38am UTC](https://discourse.numenta.org/t/why-does-ml-use-y-as-input/9659 "2022-04-25T05:38:22Z")

</div>

A ML model can be simplified as f(x) = y. For an image classifier, a typical input would be a digital image and the output would be a class. The brain on the other hand, processes (f) signals from the eye (x) and outpu…

---

## [Obstacles to widespread commercial adoption of sparsity in the ML industry?](https://discourse.numenta.org/t/obstacles-to-widespread-commercial-adoption-of-sparsity-in-the-ml-industry/9583)

<div class="topic-metadata">

**Author:** [@strangecosmos](https://discourse.numenta.org/u/strangecosmos)\
**Replies:** 38\
**Last updated:** [April 6, 2022, 1:25pm UTC](https://discourse.numenta.org/t/obstacles-to-widespread-commercial-adoption-of-sparsity-in-the-ml-industry/9583 "2022-04-06T13:25:01Z")

</div>

Numenta has shown sparsity has order(s) of magnitude effects on price-performance at least under some conditions. What obstacles remain before sparsity is widely or even universally adopted in commercial applications of …

---

## [AGI - what part does HTM/TBT play?](https://discourse.numenta.org/t/agi-what-part-does-htm-tbt-play/9445)

<div class="topic-metadata">

**Author:** [@EEProf](https://discourse.numenta.org/u/EEProf)\
**Replies:** 69\
**Last updated:** [February 23, 2022, 3:04pm UTC](https://discourse.numenta.org/t/agi-what-part-does-htm-tbt-play/9445 "2022-02-23T15:04:27Z")

</div>

Numenta, vis-à-vis Jeff Hawkins, has indeed addressed this and they are steering clear of consciousness as they should.

---

## [How do varous neural networks compare to real brains?](https://discourse.numenta.org/t/how-do-varous-neural-networks-compare-to-real-brains/9351)

<div class="topic-metadata">

**Author:** [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Replies:** 0\
**Last updated:** [January 20, 2022, 1:38pm UTC](https://discourse.numenta.org/t/how-do-varous-neural-networks-compare-to-real-brains/9351 "2022-01-20T13:38:03Z")

</div>

I have to wonder how HTM/TBT would stack up in this competition?

---

## [Is there any real world public dataset of anomaly detection example?](https://discourse.numenta.org/t/is-there-any-real-world-public-dataset-of-anomaly-detection-example/9209)

<div class="topic-metadata">

**Author:** [@Yoni\_Cohen](https://discourse.numenta.org/u/Yoni_Cohen)\
**Replies:** 3\
**Last updated:** [December 10, 2021, 11:44pm UTC](https://discourse.numenta.org/t/is-there-any-real-world-public-dataset-of-anomaly-detection-example/9209 "2021-12-10T23:44:25Z")

</div>

Hello everyone! After some serious search online I could not find any real world example of using HTM for anomaly detection. By “real world dataset” I mean a standard, public, well tested, annotated dataset - for exam…

---

## [Anomaly detection with HTM.core model on sine](https://discourse.numenta.org/t/anomaly-detection-with-htm-core-model-on-sine/8975)

<div class="topic-metadata">

**Author:** [@Robbert\_3](https://discourse.numenta.org/u/Robbert_3)\
**Replies:** 14\
**Last updated:** [December 8, 2021, 9:37am UTC](https://discourse.numenta.org/t/anomaly-detection-with-htm-core-model-on-sine/8975 "2021-12-08T09:37:54Z")

</div>

The last few weeks I have been busy with the HTM.core package to make a nice HTM model. I didn’t quite get the results I wanted on the dataset and for that reason I am now trying to make a good model on an almost perfec…

---

## [Could a GPT-3 type systems be useful in research?](https://discourse.numenta.org/t/could-a-gpt-3-type-systems-be-useful-in-research/9028)

<div class="topic-metadata">

**Author:** [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Replies:** 14\
**Last updated:** [October 9, 2021, 12:01pm UTC](https://discourse.numenta.org/t/could-a-gpt-3-type-systems-be-useful-in-research/9028 "2021-10-09T12:01:47Z")

</div>

We have had a few posts where there is some suspicion that the poster is, in fact, a GTP-3 agent. The fact that it is just a suspicion and not blatantly obvious is in itself a tip of the hat to a successful Turning test…

---

## [Model in Machine Learning and Data Science](https://discourse.numenta.org/t/model-in-machine-learning-and-data-science/9008)

<div class="topic-metadata">

**Author:** [@Arjun\_Kumar](https://discourse.numenta.org/u/Arjun_Kumar)\
**Replies:** 1\
**Last updated:** [October 3, 2021, 3:20pm UTC](https://discourse.numenta.org/t/model-in-machine-learning-and-data-science/9008 "2021-10-03T15:20:24Z")

</div>

Hello All, I want to know which one is good from model’s point of view between data science and machine learning. Yesterday I checked the difference between them from this source and As I know as model wise, In the conte…

---

## [HTM Anomaly Detection on image encodings problem](https://discourse.numenta.org/t/htm-anomaly-detection-on-image-encodings-problem/9001)

<div class="topic-metadata">

**Author:** [@vladim0105](https://discourse.numenta.org/u/vladim0105)\
**Replies:** 0\
**Last updated:** [September 30, 2021, 2:08pm UTC](https://discourse.numenta.org/t/htm-anomaly-detection-on-image-encodings-problem/9001 "2021-09-30T14:08:47Z")

</div>

Hi, I am performing some anomaly detection on image encodings, the idea being that images classified as “scorpions” will have similar latent representation to eachother in a CNN, while “cars” will have a different laten…

---

## [Machine Learning Career](https://discourse.numenta.org/t/machine-learning-career/8870)

<div class="topic-metadata">

**Author:** [@Arjun\_Kumar](https://discourse.numenta.org/u/Arjun_Kumar)\
**Replies:** 3\
**Last updated:** [August 18, 2021, 5:03am UTC](https://discourse.numenta.org/t/machine-learning-career/8870 "2021-08-18T05:03:03Z")

</div>

Hello All, I am switching from my current field which is tech writer to machine learning engineer. I want to know, Is it the right choice to start a career in this field? Can anyone explain me whats the job responsibilit…

---

## [Problem with Generalization and inversely correlated variables (or constraint variables)](https://discourse.numenta.org/t/problem-with-generalization-and-inversely-correlated-variables-or-constraint-variables/7137)

<div class="topic-metadata">

**Author:** [@Jose\_Cueto](https://discourse.numenta.org/u/Jose_Cueto)\
**Replies:** 13\
**Last updated:** [August 9, 2021, 9:12am UTC](https://discourse.numenta.org/t/problem-with-generalization-and-inversely-correlated-variables-or-constraint-variables/7137 "2021-08-09T09:12:45Z")

</div>

Question Why is ML fixated with the concept of Generalization when constraint variables/variables (e.g. weights in ANN) can be inversely correlated? Hence Generalization will be a big question, its best performance will…

---

## [DARPA TAMI (Time Aware Machine Intelligence) program](https://discourse.numenta.org/t/darpa-tami-time-aware-machine-intelligence-program/8777)

<div class="topic-metadata">

**Author:** [@JohnAlley](https://discourse.numenta.org/u/JohnAlley)\
**Replies:** 1\
**Last updated:** [July 18, 2021, 3:29pm UTC](https://discourse.numenta.org/t/darpa-tami-time-aware-machine-intelligence-program/8777 "2021-07-18T15:29:37Z")

</div>

This current iteration of the program seems to be closed, but the time intrinsic (time inextricable) nature of HTM seems to be a perfect fit for what the Defense Dept. is after. Has anyone looked into this? Pullquote: …

---

## [FNet using the Fourier Transform to mix tokens](https://discourse.numenta.org/t/fnet-using-the-fourier-transform-to-mix-tokens/8587)

<div class="topic-metadata">

**Author:** [@SeanOConnor](https://discourse.numenta.org/u/SeanOConnor)\
**Replies:** 3\
**Last updated:** [May 29, 2021, 10:11am UTC](https://discourse.numenta.org/t/fnet-using-the-fourier-transform-to-mix-tokens/8587 "2021-05-29T10:11:24Z")

</div>

There is a paper about FNet where they use the Fourier Transform to mix tokens. Using the property that a change in one input element alters all the output elements. https://youtu.be/JJR3pBl78zw The fast Walsh Hadamar…

---

## [HTM or SparseCNNs for video footage classification problem](https://discourse.numenta.org/t/htm-or-sparsecnns-for-video-footage-classification-problem/8591)

<div class="topic-metadata">

**Author:** [@rubenmoray](https://discourse.numenta.org/u/rubenmoray)\
**Replies:** 1\
**Last updated:** [May 29, 2021, 6:08am UTC](https://discourse.numenta.org/t/htm-or-sparsecnns-for-video-footage-classification-problem/8591 "2021-05-29T06:08:46Z")

</div>

Hello, this is my first post, and I hope not the last one. I’m here to ask for advice as I am starting my dissertation for a masters degree and I want to implement the HTM concept in it. I am working with videos from su…

---

## [SLIDE : In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems](https://discourse.numenta.org/t/slide-in-defense-of-smart-algorithms-over-hardware-acceleration-for-large-scale-deep-learning-systems/7278)

<div class="topic-metadata">

**Author:** [@vpuente](https://discourse.numenta.org/u/vpuente)\
**Replies:** 3\
**Last updated:** [April 16, 2021, 6:39pm UTC](https://discourse.numenta.org/t/slide-in-defense-of-smart-algorithms-over-hardware-acceleration-for-large-scale-deep-learning-systems/7278 "2021-04-16T18:39:58Z")

</div>

This might be interesting. It uses an interesting form of “sparsity” to alleviate the computational cost of DL. More info for LSH in Can be this used for a better encoder???

---

## [ML and Deep Learning to automatically create AGI?](https://discourse.numenta.org/t/ml-and-deep-learning-to-automatically-create-agi/6501)

<div class="topic-metadata">

**Author:** [@jordan.kay](https://discourse.numenta.org/u/jordan.kay)\
**Replies:** 132\
**Last updated:** [January 16, 2021, 6:26pm UTC](https://discourse.numenta.org/t/ml-and-deep-learning-to-automatically-create-agi/6501 "2021-01-16T18:26:29Z")

</div>

This thread An interesting benchmark for intelligent agents got me thinking about what it would be like to have a suite of environments that could be used as a benchmark to gauge Artificial General Intelligence. Such th…

---

## [𝐂𝐚𝐧𝐨𝐧𝐢𝐜𝐚𝐥 𝐂𝐚𝐩𝐬𝐮𝐥𝐞𝐬: 𝐔𝐧𝐬𝐮𝐩𝐞𝐫𝐯𝐢𝐬𝐞𝐝 𝐂𝐚𝐩𝐬𝐮𝐥𝐞𝐬 𝐢𝐧 𝐂𝐚𝐧𝐨𝐧𝐢𝐜𝐚𝐥 𝐏𝐨𝐬𝐞](https://discourse.numenta.org/t/topic/8177)

<div class="topic-metadata">

**Author:** [@akamburov](https://discourse.numenta.org/u/akamburov)\
**Replies:** 0\
**Last updated:** [January 15, 2021, 12:14pm UTC](https://discourse.numenta.org/t/topic/8177 "2021-01-15T12:14:48Z")

</div>

S𝐮𝐧 𝐞𝐭 𝐚𝐥. “𝐂𝐚𝐧𝐨𝐧𝐢𝐜𝐚𝐥 𝐂𝐚𝐩𝐬𝐮𝐥𝐞𝐬: 𝐔𝐧𝐬𝐮𝐩𝐞𝐫𝐯𝐢𝐬𝐞𝐝 𝐂𝐚𝐩𝐬𝐮𝐥𝐞𝐬 𝐢𝐧 𝐂𝐚𝐧𝐨𝐧𝐢𝐜𝐚𝐥 𝐏𝐨𝐬𝐞” A network design that realizes the concept of “𝐦𝐞𝐧𝐭𝐚𝐥 𝐩𝐢𝐜𝐭𝐮𝐫𝐞” for unsupervised 3D deep learning. https://canonical-capsules.github.io/ Any idea…

---

## [Can HTM extract features?](https://discourse.numenta.org/t/can-htm-extract-features/7187)

<div class="topic-metadata">

**Author:** [@way-sal](https://discourse.numenta.org/u/way-sal)\
**Replies:** 7\
**Last updated:** [December 13, 2020, 3:24pm UTC](https://discourse.numenta.org/t/can-htm-extract-features/7187 "2020-12-13T15:24:26Z")

</div>

Hi guy, how HTM can extract features (specific features) thanks.

---

## [\[Video\] Intel's update on Neuromorphic chip Loihi](https://discourse.numenta.org/t/video-intels-update-on-neuromorphic-chip-loihi/8096)

<div class="topic-metadata">

**Author:** [@maxerbubba](https://discourse.numenta.org/u/maxerbubba)\
**Replies:** 0\
**Last updated:** [December 12, 2020, 4:28am UTC](https://discourse.numenta.org/t/video-intels-update-on-neuromorphic-chip-loihi/8096 "2020-12-12T04:28:50Z")

</div>

This cheesy video references 10+ recent algorithm papers, on a spectrum from backpropagation to bio-inspired. One paper on olfactory learning caught my eye: Rapid online learning and robust recall in a neuromorphic…

---

## [Numenta Technology Demonstration: Sparse networks perform inference 50 times faster than dense networks, with competitive accuracy](https://discourse.numenta.org/t/numenta-technology-demonstration-sparse-networks-perform-inference-50-times-faster-than-dense-networks-with-competitive-accuracy/7998)

<div class="topic-metadata">

**Author:** [@cmaver](https://discourse.numenta.org/u/cmaver)\
**Replies:** 6\
**Last updated:** [November 12, 2020, 12:30am UTC](https://discourse.numenta.org/t/numenta-technology-demonstration-sparse-networks-perform-inference-50-times-faster-than-dense-networks-with-competitive-accuracy/7998 "2020-11-12T00:30:56Z")

</div>

Hi all, Today we announced a technology demonstration with results showing 50x speed-up on inference tasks in deep learning networks using sparse algorithms, with no loss in accuracy. We ran sparse and dense networks o…

---

## [Integrating HTM with Monte Carlo Tree Search](https://discourse.numenta.org/t/integrating-htm-with-monte-carlo-tree-search/7850)

<div class="topic-metadata">

**Author:** [@ayaz-amin](https://discourse.numenta.org/u/ayaz-amin)\
**Replies:** 1\
**Last updated:** [September 1, 2020, 6:04am UTC](https://discourse.numenta.org/t/integrating-htm-with-monte-carlo-tree-search/7850 "2020-09-01T06:04:05Z")

</div>

It’s as the title suggests. Is integrating HTM with such a heuristic doable, like AlphaZero / MuZero? At the moment, the CLA is not mathematically grounded so there is not much that can be done, though theoretically it s…

---

## [Applying HTM to the Omniglot dataset](https://discourse.numenta.org/t/applying-htm-to-the-omniglot-dataset/7840)

<div class="topic-metadata">

**Author:** [@ayaz-amin](https://discourse.numenta.org/u/ayaz-amin)\
**Replies:** 3\
**Last updated:** [August 31, 2020, 3:22am UTC](https://discourse.numenta.org/t/applying-htm-to-the-omniglot-dataset/7840 "2020-08-31T03:22:38Z")

</div>

So I recently applied HTM to the one shot learning problem of the Omniglot challenge. Let’s just say it did not go too well. Perhaps HTM is not well suited towards image data? Or maybe my implementation had some silly m…

---

## [Neural networks are moving out to the the edge!](https://discourse.numenta.org/t/neural-networks-are-moving-out-to-the-the-edge/7820)

<div class="topic-metadata">

**Author:** [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Replies:** 2\
**Last updated:** [August 25, 2020, 9:13pm UTC](https://discourse.numenta.org/t/neural-networks-are-moving-out-to-the-the-edge/7820 "2020-08-25T21:13:37Z")

</div>

Perhaps the advantages of HTM have a place in this market. Machine learning is a rapidly growing field, especially as deep learning moves to the edge and more and more engineers build applications that include some form…

---

## [Interesting A(G)I attempt](https://discourse.numenta.org/t/interesting-a-g-i-attempt/7792)

<div class="topic-metadata">

**Author:** [@cezar\_t](https://discourse.numenta.org/u/cezar_t)\
**Replies:** 2\
**Last updated:** [August 12, 2020, 4:47pm UTC](https://discourse.numenta.org/t/interesting-a-g-i-attempt/7792 "2020-08-12T16:47:27Z")

</div>

I just found this Russian group which also approaches AI by making assumptions on the underlying processing in the brain with emphasis on minicolumns functions and role. Some of their points might be weird but others a…

---

## [Should we do any Tensorflow 2.0 development?](https://discourse.numenta.org/t/should-we-do-any-tensorflow-2-0-development/5757)

<div class="topic-metadata">

**Author:** [@SimLeek](https://discourse.numenta.org/u/SimLeek)\
**Replies:** 5\
**Last updated:** [June 11, 2020, 5:09pm UTC](https://discourse.numenta.org/t/should-we-do-any-tensorflow-2-0-development/5757 "2020-06-11T17:09:32Z")

</div>

Someone mentioned Tensorflow 2.0 and Keras in the last hacker’s hangout, so I wondered: could there be a good enough reason to move to it from PyTorch? Has anyone looked into the benefits of tensorflow 2.0 yet? Can it co…

[Previous page](https://discourse.numenta.org/c/engineering/machine-learning/48.md)

[Next page](https://discourse.numenta.org/c/engineering/machine-learning/48.md?page=2)