A Theory of How Columns in the Neocortex Enable Learning the Structure of the World.
Jeff Hawkins, Subutai Ahmad, Yuwei Cui
July 2018 Front Neural CircuitsSynopsis of Social media discussions
Discussions include enthusiastic endorsements such as calling it 'super cool' and emphasizing its relevance to AI and brain understanding. Phrases like 'good weekend reading' and mentions of the potential for future AI applications show excitement and recognition of the publication’s significance. Technical references, such as the voting mechanism in cortical columns and the comparison to neural transforms, demonstrate deep engagement and a recognition of the research's potential impact.
Agreement
Moderate agreementMost discussions support or accept the significance of the research, with some referencing it as foundational or important for understanding brain functions.
Interest
High level of interestPosts demonstrate high curiosity, often highlighting how the publication relates to broader topics like AI, neuroscience, and cognitive models, indicating strong engagement.
Engagement
High engagementSeveral comments delve into detailed aspects such as the voting mechanisms between cortical columns, implications for AI, and neural modeling, reflecting deep engagement.
Impact
Moderate level of impactWhile many acknowledge the research's importance, few claim it has immediate practical impact, suggesting a perception of long-term influence rather than urgent change.
Social Mentions
YouTube
8 Videos
3 Posts
39 Posts
News
18 Articles
2 Posts
Metrics
Video Views
214,763
Total Likes
9,324
Extended Reach
448,491
Social Features
70
Timeline: Posts about article
Top Social Media Posts
Vedat
Muazzam bir araştırma. Kendi evreninde, kendi oyununda, kendi yansımalarının tuzağından kurtulabilmenin yolu.
https://t.co/Pa73uc9sdx
view full post
Numenta
In the continuation of last week’s research meeting, @SubutaiAhmad explains voting in the Thousand Brains Theory. In part two, he explains how cortical columns vote across multiple columns.
TLDW below
view full post
Numenta
In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_eliav et al., and our researcher Ben Cohen reviewed a paper on representational drift by Schoonover et al.
https://t.co/xo4RcJI0To
view full post
Numenta
Read more about how cortical columns in the neocortex enable learning the structure of the world here:
https://t.co/CWAfiX0aLx
view full post
Subutai Ahmad
@IntuitMachine @Numenta Thanks! We published a detailed model in 2017 of how voting might work using long-range lateral connections that exist between cortical columns (layer 2).
https://t.co/8KQqpmOvtI
view full post
Posts referencing the article
Neocortical Columns and Object Recognition in the Brain
This video explores how neocortical columns enable learning and recognition of complex objects by integrating sensory input over time and space, emphasizing the role of columns in perception and understanding beyond individual receptive fields.
December 2, 2021
146,573 views
Role of Allocentric Location in Neocortex Function
In this talk, Jeff Hawkins describes a theory that the neocortex processes both sensory data and allocentric location representations. This dual input allows cortical columns to build comprehensive object models, enhancing our understanding of neocortical function and memory.
December 15, 2017
32,888 views
Neocortical Columns and Object Learning in the Brain
This video accompanies the paper titled A Theory of How Columns in the Neocortex Enable Learning the Structure of the World by Jeff Hawkins, Subutai Ahmad, and Yuwei Cui. It explores how cortical columns use location information to build predictive models of objects, applicable across sensory modalities, including touch.
October 25, 2017
16,700 views
Understanding Active Dendrites in Neuron Prediction and Context Integration
Subutai Ahmad reviews the biology behind active dendrites and discusses how they influence neuronal computation. He explores various findings on dendritic functions, emphasizing their role in prediction and context integration in neural networks, supported by key scientific studies.
August 23, 2021
5,375 views
Understanding Cortical Column Voting Mechanisms in Neural Networks
Subutai Ahmad explains the voting mechanisms in cortical columns developed by Numenta, highlighting how neocortical columns facilitate learning and object recognition by integrating sensory input over time and space.
April 14, 2022
5,007 views
Voting Mechanisms in the Thousand Brains Model of Cortical Columns
Subutai Ahmad explains how cortical columns use voting mechanisms to learn predictive models of sensorimotor sequences, laying the foundation for understanding how the neocortex recognizes complex objects through sensorimotor integration.
August 31, 2021
3,386 views
Thousand Brains Theory: Hierarchical Voting for Rapid Object Recognition
In this video, Subutai Ahmad explains how the neocortex uses hierarchy to vote across levels and sensory modalities, enabling rapid consensus on objects through multiple cortical columns and sparse connections, leading to faster inference and better perception.
September 3, 2021
2,490 views
Multiscale Neural Representation and Drift in the Neocortex
This video reviews key research on neuronal encoding of spatial perception in the hippocampus of flying bats and stability of odor responses in the olfactory cortex. It also explores how neocortical columns enable object recognition through integrated sensory input.
July 16, 2021
2,346 views
-
RT @vedattozyilmaz: Muazzam bir araştırma. Kendi evreninde, kendi oyununda, kendi yansımalarının tuzağından kurtulabilmenin yolu. https://…
view full postMarch 2, 2025
2
-
Vedat
@vedattozyilmaz (Twitter)Muazzam bir araştırma. Kendi evreninde, kendi oyununda, kendi yansımalarının tuzağından kurtulabilmenin yolu. https://t.co/Pa73uc9sdx
view full postMarch 1, 2025
27
2
-
Quantengeist
@Light_Cynapse (Twitter)The solution is already quite clear: https://t.co/mKYMcNBCqN https://t.co/eWyRKqDQ7G https://t.co/ikdMjts9TC
view full postFebruary 13, 2025
-
Yu Kanazawa
@knzw783 (Twitter)Hawkins, J., Ahmad, S., & Cui, Y. (2017). A theory of how columns in the neocortex enable learning the structure of the world. Frontiers in Neural Circuits, 11, 81. https://t.co/AJd49ITYdE
view full postJune 24, 2024
1
-
Hidenori Goto
@hidenorigoto (Twitter)https://t.co/vpUP9SwONd
view full postJune 2, 2024
-
Luca Dellanna
@DellAnnaLuca (Twitter)@goldstein_aa These are the two most important papers IMHO: https://t.co/5y0UJk79T1 https://t.co/Qn4eA3wvzq
view full postApril 11, 2024
2
-
Luca Dellanna
@DellAnnaLuca (Twitter)@HansaMD_ It’s not the five fingers who vote. It’s the cortical columns associated to the fingers. I suggest you read this paper for a full description of how the system works: https://t.co/Qn4eA3wvzq
view full postDecember 9, 2023
1
-
Harry
@education_24x7 (Twitter)RT @hwingo: A Theory of How Columns in the #Neocortex Enable Learning the Structure of the World https://t.co/TG7kd2Ojj6 #ICYMI #neurosci…
view full postJuly 6, 2022
1
-
hwingo
@hwingo (Twitter)A Theory of How Columns in the #Neocortex Enable Learning the Structure of the World https://t.co/TG7kd2wHUw #ICYMI #neuroscience #AI #ArtificialIntelligence #brain @Numenta
view full postJuly 6, 2022
-
hwingo
@hwingo (Twitter)A Theory of How Columns in the #Neocortex Enable Learning the Structure of the World https://t.co/TG7kd2Ojj6 #ICYMI #neuroscience #AI #ArtificialIntelligence #brain @Numenta https://t.co/FXZcCEsWuJ
view full postJuly 5, 2022
1
1
-
justindegarbo
@justindegarbo (Twitter)https://t.co/7imINNVMFg
view full postJune 30, 2022
1
-
ideaXme™
@ideaxm (Twitter)RT @Numenta: Read more about how cortical columns in the neocortex enable learning the structure of the world here: https://t.co/CWAfiX0aLx
view full postMarch 12, 2022
2
-
John F Halter
@johnfhalt (Twitter)RT @Numenta: Read more about how cortical columns in the neocortex enable learning the structure of the world here: https://t.co/CWAfiX0aLx
view full postFebruary 21, 2022
2
-
Numenta
@Numenta (Twitter)Read more about how cortical columns in the neocortex enable learning the structure of the world here: https://t.co/CWAfiX0aLx
view full postFebruary 18, 2022
8
2
-
Doge Park
@2021DogePark (Twitter)“A Theory of How Columns in the Neocortex Enable Learning the Structure of the World.”
view full postJanuary 14, 2022
-
David Kedmey
@DKedmey (Twitter)2/4 https://t.co/WZPnuAPYIX
view full postSeptember 22, 2021
-
Numenta
@Numenta (Twitter)For reference: "Columns" paper: https://t.co/sjN1hzZ8Oz "Frameworks" Paper: https://t.co/g7JL2KDG3h
view full postSeptember 20, 2021
6
-
Pedro Coelho Silva
@PedroCoSilva (Twitter)RT @Numenta: In the next research meeting, he will review voting across multiple columns. Read our "A Theory of How Columns in the Neocor…
view full postSeptember 11, 2021
1
-
Rubén Martínez
@EldarSilver (Twitter)RT @Numenta: In the continuation of last week’s research meeting, @SubutaiAhmad explains voting in the Thousand Brains Theory. In part two,…
view full postSeptember 4, 2021
4
-
RyuZoku ObjectiveReality MatrixRed
@Ryu_Zoku (Twitter)RT @Numenta: In the continuation of last week’s research meeting, @SubutaiAhmad explains voting in the Thousand Brains Theory. In part two,…
view full postSeptember 3, 2021
4
-
Stephen Pimentel
@StephenPiment (Twitter)RT @Numenta: In the continuation of last week’s research meeting, @SubutaiAhmad explains voting in the Thousand Brains Theory. In part two,…
view full postSeptember 3, 2021
4
-
Numenta
@Numenta (Twitter)In the continuation of last week’s research meeting, @SubutaiAhmad explains voting in the Thousand Brains Theory. In part two, he explains how cortical columns vote across multiple columns. TLDW below
view full postSeptember 3, 2021
19
4
-
Rommel Infante Asto
@R_InfanteAsto (Twitter)miles de modelados del mundo de forma simultanea, que se complementan y compiten entre sí, hasta llegar a un estado final de decisión a través de un mecanismo de "votación neuronal"
view full postAugust 13, 2021
2
-
ngszehing
@ngszehin (Twitter)This is a good weekend reading, to understand how the future of AI might work by understanding how human’s brain works: A Theory of How Columns in the Neocortex Enable Learning the Structure of the World https://t.co/04XGZJEPJ4
view full postAugust 7, 2021
-
Tamir Eliav
@tamir_eliav (Twitter)RT @Numenta: In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_elia…
view full postJuly 28, 2021
3
-
Viviane Clay
@vkakerbeck (Twitter)RT @Numenta: In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_elia…
view full postJuly 19, 2021
3
-
Andy Xor
@andy_xor (Twitter)RT @Numenta: In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_elia…
view full postJuly 16, 2021
3
-
Subutai Ahmad
@SubutaiAhmad (Twitter)RT @Numenta: In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_elia…
view full postJuly 16, 2021
3
-
Numenta
@Numenta (Twitter)In this meeting, Numenta intern Jack Schenkman reviewed a paper on multiscale representation in the hippocampus by @tamir_eliav et al., and our researcher Ben Cohen reviewed a paper on representational drift by Schoonover et al. https://t.co/xo4RcJI0To
view full postJuly 16, 2021
19
3
-
Subutai Ahmad
@SubutaiAhmad (Twitter)@IntuitMachine @Numenta Thanks! We published a detailed model in 2017 of how voting might work using long-range lateral connections that exist between cortical columns (layer 2). https://t.co/8KQqpmOvtI
view full postMarch 5, 2021
7
-
Chomba Bupe
@ChombaBupe (Twitter)@connectedregio1 @RebelScience @markcannon5 For example this is a very complicated paper https://t.co/gvmJDUNHko
view full postFebruary 29, 2020
2
-
@rhyolight (Twitter)@bbc1183 @mrcslws We have experimented with union pooling during sensorimotor inference. In what we call the "output layer" of this paper we are pooling objects in a union. https://t.co/lpvL3zKPFN
view full postOctober 31, 2019
1
-
Ben Hoyle
@bjh_ip (Twitter)A Theory of How Columns in the Neocortex Enable Learning the Structure of the World [Reading this through I think the example of vision provides a stronger argument than sensorimotor sensing, e.g. considering the constant changing information on the fovea] https://t.co/V7tajeiax9
view full postOctober 2, 2019
-
@art3starr@mastodon.social : ☮️ : starrgirl.eth
@art3starr (Twitter)I also mentioned @Numenta ‘s research on reverse-engineering the neocortex, when @jfcloutier asked about universal function... here’s a short video that explains the magic of allocentric location. Super cool stuff: https://t.co/9Q4wfvgMC1
view full postSeptember 21, 2019
1
-
sahz
@0xhzsa (Twitter)https://t.co/wzmrTEnLyn
view full postAugust 10, 2019
-
Chomba Bupe
@ChombaBupe (Twitter)HTM neurons in the output layer seem to doing hough transform. https://t.co/gvmJDUNHko
view full postJuly 30, 2019
-
Postscapes
@postscapes (Twitter)@el33th4xor Any cross-over potential in how Avalanche protocol works in the same way as neocortex in this view? https://t.co/aPrPtRLuZD
view full postJuly 10, 2019
-
MohamadSama7
@Sama7Mohamad (Twitter)RT @Numenta: .@CDMaver debunks the belief that the brain fills in the gap in our vision to compensate for our blind spot using our sensorim…
view full postFebruary 10, 2019
6
-
Mathew Abonyi
@mabonyi (Twitter)If this is the case and everything springs from loci, wouldn’t it mean tests of intelligence could be boiled down to a person’s aptitude at primitive spatial awareness, memory, and composition? https://t.co/HcNFLq9yhL
view full postJanuary 14, 2019
Abstract Synopsis
- The article discusses how neocortical columns, organized into layers, facilitate learning and recognizing complex objects by integrating sensory input over time and space.
- It introduces a network model where each column computes a location relative to an object, which, combined with sensory data, improves object recognition and allows understanding of objects extending beyond individual cells’ receptive fields.
- Simulations indicate that small, multi-column networks can efficiently learn to recognize many objects with minimal movements, suggesting that the columnar structure of the cortex plays a powerful role in perception and modeling beyond previous assumptions.]
AyasuAbray
@AyasuAbray (Twitter)