# \#sdrs

**URL:** https://discourse.numenta.org/tag/sdrs/24.md

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

---

## [Capacity of SDR with overlap formula](https://discourse.numenta.org/t/capacity-of-sdr-with-overlap-formula/11071)

<div class="topic-metadata">

**Author:** [@ivan256](https://discourse.numenta.org/u/ivan256)\
**Replies:** 2\
**Last updated:** [November 2, 2023, 9:55pm UTC](https://discourse.numenta.org/t/capacity-of-sdr-with-overlap-formula/11071 "2023-11-02T21:55:52Z")

</div>

SDR capacity formula is just Combination. But it is true if we consider 2 SDRs are different even if they have merely one different bit. In practice we considers two SDRs are different if they have for example 20% of di…

---

## [SDRs and/or Graphs - modelling the data structures of the brain](https://discourse.numenta.org/t/sdrs-and-or-graphs-modelling-the-data-structures-of-the-brain/5326)

<div class="topic-metadata">

**Author:** [@Curious\_Mike](https://discourse.numenta.org/u/Curious_Mike)\
**Replies:** 2\
**Last updated:** [February 14, 2022, 9:35pm UTC](https://discourse.numenta.org/t/sdrs-and-or-graphs-modelling-the-data-structures-of-the-brain/5326 "2022-02-14T21:35:54Z")

</div>

Hi All, am just looking for a steer to any discussion on this forum, or related, covering the questions: How close are SDRs to “Graphs” (static or dynamic or not at all)? What can, or has, HTM leveraged from Graph the…

---

## [Decoding in HTM](https://discourse.numenta.org/t/decoding-in-htm/9173)

<div class="topic-metadata">

**Author:** [@hodor-cna](https://discourse.numenta.org/u/hodor-cna)\
**Replies:** 1\
**Last updated:** [December 1, 2021, 9:51pm UTC](https://discourse.numenta.org/t/decoding-in-htm/9173 "2021-12-01T21:51:02Z")

</div>

How does HTM decode the SDRS after temporal poolering?

---

## [SDR math regarding scent detection in mixtures and dog noses](https://discourse.numenta.org/t/sdr-math-regarding-scent-detection-in-mixtures-and-dog-noses/8219)

<div class="topic-metadata">

**Author:** [@Julius](https://discourse.numenta.org/u/Julius)\
**Replies:** 1\
**Last updated:** [February 3, 2021, 7:29am UTC](https://discourse.numenta.org/t/sdr-math-regarding-scent-detection-in-mixtures-and-dog-noses/8219 "2021-02-03T07:29:52Z")

</div>

Hi, I thought it would be interesting to discuss how SDR’s influence the number of patterns we can detect/predict/reconstruct in a single SDR, in this case how two different patterns scents vanilla and lilac when combine…

---

## [Question on desired sparsity](https://discourse.numenta.org/t/question-on-desired-sparsity/6901)

<div class="topic-metadata">

**Author:** [@helena\_Thielen](https://discourse.numenta.org/u/helena_Thielen)\
**Replies:** 4\
**Last updated:** [December 11, 2019, 12:35am UTC](https://discourse.numenta.org/t/question-on-desired-sparsity/6901 "2019-12-11T00:35:56Z")

</div>

Hey everybody, I have a question about the desired sparsity. Well, typically we want 2%. Now, in the SP algorithm (lets assume we have global inhibition) we have 2% of the columns active. So lets look at an example: …

---

## [Introducing new project: image2sdr](https://discourse.numenta.org/t/introducing-new-project-image2sdr/4544)

<div class="topic-metadata">

**Author:** [@Paul\_Lamb](https://discourse.numenta.org/u/Paul_Lamb)\
**Replies:** 41\
**Last updated:** [November 23, 2019, 4:01pm UTC](https://discourse.numenta.org/t/introducing-new-project-image2sdr/4544 "2019-11-23T16:01:10Z")

</div>

I have uploaded a new project on GitHub called image2sdr. It is a simple NodeJS service which you can post an image file to, and it returns an SDR. Behind the scenes what it is doing is: Classify image via Clarifai a…

---

## [Fuzzy logic, quantum computing, HTM SDRs](https://discourse.numenta.org/t/fuzzy-logic-quantum-computing-htm-sdrs/6613)

<div class="topic-metadata">

**Author:** [@breznak](https://discourse.numenta.org/u/breznak)\
**Replies:** 0\
**Last updated:** [September 20, 2019, 12:39pm UTC](https://discourse.numenta.org/t/fuzzy-logic-quantum-computing-htm-sdrs/6613 "2019-09-20T12:39:57Z")

</div>

FYI the link on using p-bits (probabilities) in HW, instead of q-bits (quantum). I wan to think about HTM sparse representations (SDRs) being able to mimic such “quantum computing”? noise robustness union of concept…

---

## [Classification and Recall With Binary Hyperdimensional Computing: Tradeoffs in Choice of Density and Mapping Characteristics](https://discourse.numenta.org/t/classification-and-recall-with-binary-hyperdimensional-computing-tradeoffs-in-choice-of-density-and-mapping-characteristics/3903)

<div class="topic-metadata">

**Author:** [@bkutt](https://discourse.numenta.org/u/bkutt)\
**Replies:** 0\
**Last updated:** [May 14, 2018, 7:39pm UTC](https://discourse.numenta.org/t/classification-and-recall-with-binary-hyperdimensional-computing-tradeoffs-in-choice-of-density-and-mapping-characteristics/3903 "2018-05-14T19:39:18Z")

</div>

An interesting comparative study in the efficacy of sparse and dense binary vector representations. https://ieeexplore.ieee.org/document/8331890/

---

## [A rough idea about explanation of Déjà vu and dream](https://discourse.numenta.org/t/a-rough-idea-about-explanation-of-deja-vu-and-dream/5835)

<div class="topic-metadata">

**Author:** [@root](https://discourse.numenta.org/u/root)\
**Replies:** 2\
**Last updated:** [April 16, 2019, 7:41am UTC](https://discourse.numenta.org/t/a-rough-idea-about-explanation-of-deja-vu-and-dream/5835 "2019-04-16T07:41:00Z")

</div>

I think that HTM Theory can explan the déjà . Maybe in high level neocortex, SDR’s demension is not very big. because the probability of false positive(experience of déjà vu) is not so small.

---

## [Deep learning meets sparse distributed representations: The Kanerva Machine](https://discourse.numenta.org/t/deep-learning-meets-sparse-distributed-representations-the-kanerva-machine/3743)

<div class="topic-metadata">

**Author:** [@jakebruce](https://discourse.numenta.org/u/jakebruce)\
**Replies:** 7\
**Last updated:** [March 23, 2019, 1:22am UTC](https://discourse.numenta.org/t/deep-learning-meets-sparse-distributed-representations-the-kanerva-machine/3743 "2019-03-23T01:22:01Z")

</div>

“We present an end-to-end trained memory system that quickly adapts to new data and generates samples like them. Inspired by Kanerva’s sparse distributed memory, it has a robust distributed reading and writing mechanism.…

---

## [Initial real time Vision to SDR encoder](https://discourse.numenta.org/t/initial-real-time-vision-to-sdr-encoder/5000)

<div class="topic-metadata">

**Author:** [@SimLeek](https://discourse.numenta.org/u/SimLeek)\
**Replies:** 10\
**Last updated:** [January 21, 2019, 4:08am UTC](https://discourse.numenta.org/t/initial-real-time-vision-to-sdr-encoder/5000 "2019-01-21T04:08:31Z")

</div>

Well it’s been a while. I’ve been pretty busy, but whenever I felt like I had enough spare time, I’ve continued working on something inspired by this forum: This uses OpenCV and TensorFlow to translate rgb camera input…

---

## [Robustness of SDR](https://discourse.numenta.org/t/robustness-of-sdr/5115)

<div class="topic-metadata">

**Author:** [@JosefOndrej](https://discourse.numenta.org/u/JosefOndrej)\
**Replies:** 7\
**Last updated:** [January 17, 2019, 6:03pm UTC](https://discourse.numenta.org/t/robustness-of-sdr/5115 "2019-01-17T18:03:53Z")

</div>

In HTM school and the paper about properties of SDRs the robustness of SDRs is demonstrated with respect to switching some % of active bits. Why is this the relevant metric? Is there some neurological explanation? Why no…

---

## [2d vs 3d vs nd SDRs](https://discourse.numenta.org/t/2d-vs-3d-vs-nd-sdrs/3070)

<div class="topic-metadata">

**Author:** [@JustAnEngineerO](https://discourse.numenta.org/u/JustAnEngineerO)\
**Replies:** 4\
**Last updated:** [November 8, 2018, 12:31pm UTC](https://discourse.numenta.org/t/2d-vs-3d-vs-nd-sdrs/3070 "2018-11-08T12:31:09Z")

</div>

Quick question, I’ve watched all the HTM school videos. I noticed in all the ideas of the SDRs, especially the union videos, the focus is on 2-dimensional SDRs, why not 3d or n-dimensional SDRs? Was there a specific reas…

---

## [Why the overlap set of SDR isn't w choose b?](https://discourse.numenta.org/t/why-the-overlap-set-of-sdr-isnt-w-choose-b/4782)

<div class="topic-metadata">

**Author:** [@OhMyHTM](https://discourse.numenta.org/u/OhMyHTM)\
**Replies:** 2\
**Last updated:** [October 22, 2018, 4:44am UTC](https://discourse.numenta.org/t/why-the-overlap-set-of-sdr-isnt-w-choose-b/4782 "2018-10-22T04:44:54Z")

</div>

i’m learning SDR overlap set at HTM School. give a SDR “x”, x is one of all encodings, so this SDRx is unique and fixed, right? so, the number of all SDRs with b bits overlaps with SDRx is w choose b, am i right? w is…

---

## [Graded (non-binary) SDRs](https://discourse.numenta.org/t/graded-non-binary-sdrs/3776)

<div class="topic-metadata">

**Author:** [@sebjwallace](https://discourse.numenta.org/u/sebjwallace)\
**Replies:** 4\
**Last updated:** [April 13, 2018, 9:44pm UTC](https://discourse.numenta.org/t/graded-non-binary-sdrs/3776 "2018-04-13T21:44:38Z")

</div>

As far as I am aware, SDRs have only been discussed as binary arrays. That makes sense when you measure activity in a single moment. However if you measure activity over a period of time the SDRs could have a frequency. …

---

## [Trading byte size for noise](https://discourse.numenta.org/t/trading-byte-size-for-noise/3514)

<div class="topic-metadata">

**Author:** [@Paul\_Lamb](https://discourse.numenta.org/u/Paul_Lamb)\
**Replies:** 18\
**Last updated:** [February 26, 2018, 11:23am UTC](https://discourse.numenta.org/t/trading-byte-size-for-noise/3514 "2018-02-26T11:23:51Z")

</div>

In a system where size in bytes of an SDR is of concern (such as synchronizing between multiple threads, or transmitting over a network), the thought occurred to me that indices to the 1 bits could be relative to each ot…

---

## [How can I reconstruct the input using an SDR?](https://discourse.numenta.org/t/how-can-i-reconstruct-the-input-using-an-sdr/3142)

<div class="topic-metadata">

**Author:** [@ivansiiito](https://discourse.numenta.org/u/ivansiiito)\
**Replies:** 13\
**Last updated:** [January 19, 2018, 12:14pm UTC](https://discourse.numenta.org/t/how-can-i-reconstruct-the-input-using-an-sdr/3142 "2018-01-19T12:14:15Z")

</div>

Hi! I have been trying to find a way so I can reconstruct the input signal out of the SDRs that it generates, but I haven’t found it so far. I mean, using the method SpatialPooler.compute(), I can get the SDR that repre…

---

## [Do we have performance numbers for SDR library?](https://discourse.numenta.org/t/do-we-have-performance-numbers-for-sdr-library/2476)

<div class="topic-metadata">

**Author:** [@pgavazzi](https://discourse.numenta.org/u/pgavazzi)\
**Replies:** 7\
**Last updated:** [January 11, 2018, 2:43pm UTC](https://discourse.numenta.org/t/do-we-have-performance-numbers-for-sdr-library/2476 "2018-01-11T14:43:15Z")

</div>

Hi, I was wondering if anybody in the community or at Numenta had done any recent performance numbers only on the SDR library as far as creating unions, checking agains’t a union, or other SDR operation with 1000s, 1M o…

---

## [Lexicon Induction From Temporal Structure](https://discourse.numenta.org/t/lexicon-induction-from-temporal-structure/3144)

<div class="topic-metadata">

**Author:** [@James\_Bowery](https://discourse.numenta.org/u/James_Bowery)\
**Replies:** 44\
**Last updated:** [January 4, 2018, 1:06pm UTC](https://discourse.numenta.org/t/lexicon-induction-from-temporal-structure/3144 "2018-01-04T13:06:04Z")

</div>

A stream of bits, about which one knows only that it is comprised of some unknown lexicon tokenized as n-bits where n is also unknown, may be thought of as a temporal structure. In other words, the HTM’s challenge is to…

---

## [Francisco explains HTM on TWiML podcast](https://discourse.numenta.org/t/francisco-explains-htm-on-twiml-podcast/2886)

<div class="topic-metadata">

**Author:** [@rhyolight](https://discourse.numenta.org/u/rhyolight)\
**Replies:** 1\
**Last updated:** [December 17, 2017, 10:33pm UTC](https://discourse.numenta.org/t/francisco-explains-htm-on-twiml-podcast/2886 "2017-12-17T22:33:59Z")

</div>

I really enjoyed this interview with Francisco:

---

## [SDR representation](https://discourse.numenta.org/t/sdr-representation/2996)

<div class="topic-metadata">

**Author:** [@zahir\_hamroune](https://discourse.numenta.org/u/zahir_hamroune)\
**Replies:** 19\
**Last updated:** [October 31, 2017, 5:23pm UTC](https://discourse.numenta.org/t/sdr-representation/2996 "2017-10-31T17:23:12Z")

</div>

Hi, I want to understand, how to pass from SDR representation (as array of 0/1) to “Matrix” view representation, I cannot see the relation between the Int Array (0…1) and the matrix representation (32x32 columns for ex…

---

## [Maximum overlap between the SDRs of two distant values](https://discourse.numenta.org/t/maximum-overlap-between-the-sdrs-of-two-distant-values/2742)

<div class="topic-metadata">

**Author:** [@caesar](https://discourse.numenta.org/u/caesar)\
**Replies:** 2\
**Last updated:** [August 25, 2017, 6:18pm UTC](https://discourse.numenta.org/t/maximum-overlap-between-the-sdrs-of-two-distant-values/2742 "2017-08-25T18:18:55Z")

</div>

Hi! I have a question related to the maximum overlap of the SDRs of two distant values (Ideally 0). In the source code of the Random Distributed Scalar Encoder (RDSE) appears the following: Dissimilar scalars should …

---

## [Sparse Distributed Representations Quiz](https://discourse.numenta.org/t/sparse-distributed-representations-quiz/2151)

<div class="topic-metadata">

**Author:** [@subutai](https://discourse.numenta.org/u/subutai)\
**Replies:** 0\
**Last updated:** [April 6, 2017, 10:35pm UTC](https://discourse.numenta.org/t/sparse-distributed-representations-quiz/2151 "2017-04-06T22:35:19Z")

</div>

Useful reference: Properties of Sparse Distributed Representations What are disadvantages of SDR's? What does the "distributed" part of SDR mean? Why is sparsity necessary? How sparse do representations need to be? Ca…

---

## [Sparse Distributed Representations](https://discourse.numenta.org/t/sparse-distributed-representations/2150)

<div class="topic-metadata">

**Author:** [@subutai](https://discourse.numenta.org/u/subutai)\
**Replies:** 0\
**Last updated:** [April 6, 2017, 10:34pm UTC](https://discourse.numenta.org/t/sparse-distributed-representations/2150 "2017-04-06T22:34:13Z")

</div>

Sparse Distributed Representations (SDRs) The Brain’s Data Structure Computers store information in “dense” representations such as a 32 bit word where all combinations of 1s and 0s are possible. By contrast, brains …

---

## [Can we deduce a Universal Encoder algorithm via the Fundamentals?](https://discourse.numenta.org/t/can-we-deduce-a-universal-encoder-algorithm-via-the-fundamentals/2080)

<div class="topic-metadata">

**Author:** [@jordan.kay](https://discourse.numenta.org/u/jordan.kay)\
**Replies:** 28\
**Last updated:** [April 4, 2017, 4:26pm UTC](https://discourse.numenta.org/t/can-we-deduce-a-universal-encoder-algorithm-via-the-fundamentals/2080 "2017-04-04T16:26:30Z")

</div>

Before we talk about a possible design for a universal encoder I think it would be best if we are on the same page. I may not understand something fundamental about Encoders, so let's take it slow... Here's what I unde…

---

## [Further reading on SDRs and Temporal Memory](https://discourse.numenta.org/t/further-reading-on-sdrs-and-temporal-memory/2034)

<div class="topic-metadata">

**Author:** [@sebjwallace](https://discourse.numenta.org/u/sebjwallace)\
**Replies:** 3\
**Last updated:** [March 24, 2017, 7:17pm UTC](https://discourse.numenta.org/t/further-reading-on-sdrs-and-temporal-memory/2034 "2017-03-24T19:17:40Z")

</div>

I came across a book from Pentti Kanerva titled Sparse Distributed Memory. It arrived today and upon looking through it I became surprised that I’d not heard of Kanerva’s work before. As the cortex is a sparse distribut…

---

## ['Cognitive Computing' using HTM](https://discourse.numenta.org/t/cognitive-computing-using-htm/1479)

<div class="topic-metadata">

**Author:** [@sebjwallace](https://discourse.numenta.org/u/sebjwallace)\
**Replies:** 0\
**Last updated:** [October 9, 2016, 1:13pm UTC](https://discourse.numenta.org/t/cognitive-computing-using-htm/1479 "2016-10-09T13:13:05Z")

</div>

A lot of marketing terms have been thrown around recently in Artificial Intelligence, one of them being 'Cognitive Computing', which implies that the computing works on the similar principles of the cortex. As far as we'…

---

## [HTM School Episode 4: SDR Sets & Unions](https://discourse.numenta.org/t/htm-school-episode-4-sdr-sets-unions/455)

<div class="topic-metadata">

**Author:** [@rhyolight](https://discourse.numenta.org/u/rhyolight)\
**Replies:** 13\
**Last updated:** [May 25, 2016, 9:22pm UTC](https://discourse.numenta.org/t/htm-school-episode-4-sdr-sets-unions/455 "2016-05-25T21:22:28Z")

</div>

https://www.youtube.com/watch?v=8WIzIBaLXIs What should I do next? Encoders Spatial Pooling 0voters Show results

---

## [The art of SDRs](https://discourse.numenta.org/t/the-art-of-sdrs/266)

<div class="topic-metadata">

**Author:** [@karchie](https://discourse.numenta.org/u/karchie)\
**Replies:** 17\
**Last updated:** [May 18, 2016, 7:53am UTC](https://discourse.numenta.org/t/the-art-of-sdrs/266 "2016-05-18T07:53:27Z")

</div>

The discussion over here got me started on one of my HTM rants. It's possible that I've aired variations of this on the mailing list before: It's both one of my biggest problems with HTM and yet not exactly a problem wit…