Limits...
Towards a mathematical theory of cortical micro-circuits.

George D, Hawkins J - PLoS Comput. Biol. (2009)

Bottom Line: Anatomical data provide a contrasting set of organizational constraints.The combination of these two constraints suggests a theoretically derived interpretation for many anatomical and physiological features and predicts several others.We also discuss how the theory and the circuit can be extended to explain cortical features that are not explained by the current model and describe testable predictions that can be derived from the model.

View Article: PubMed Central - PubMed

Affiliation: Numenta Inc., Redwood City, California, United States of America. dgeorge@numenta.com

ABSTRACT
The theoretical setting of hierarchical Bayesian inference is gaining acceptance as a framework for understanding cortical computation. In this paper, we describe how Bayesian belief propagation in a spatio-temporal hierarchical model, called Hierarchical Temporal Memory (HTM), can lead to a mathematical model for cortical circuits. An HTM node is abstracted using a coincidence detector and a mixture of Markov chains. Bayesian belief propagation equations for such an HTM node define a set of functional constraints for a neuronal implementation. Anatomical data provide a contrasting set of organizational constraints. The combination of these two constraints suggests a theoretically derived interpretation for many anatomical and physiological features and predicts several others. We describe the pattern recognition capabilities of HTM networks and demonstrate the application of the derived circuits for modeling the subjective contour effect. We also discuss how the theory and the circuit can be extended to explain cortical features that are not explained by the current model and describe testable predictions that can be derived from the model.

Show MeSH

Related in: MedlinePlus

Examples of training and testing images for an HTM network trained for visual object recognition.The top two rows are examples of training images. The bottom three rows are examples of correctly recognized test images. The last row shows test images that incorporated distracter backgrounds.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2749218&req=5

pcbi-1000532-g013: Examples of training and testing images for an HTM network trained for visual object recognition.The top two rows are examples of training images. The bottom three rows are examples of correctly recognized test images. The last row shows test images that incorporated distracter backgrounds.

Mentions: We used Numenta's NuPIC software environment to train a visual pattern recognition HTM network on which we tested the subjective contour effect. We started with an HTM network that was trained to recognize four different categories of objects: binoculars, cars, cell phones, and rubber ducks. This network had a three level HTM hierarchy. Figure 13 shows examples of training and testing images for these categories. When presented with a test image, the output from the top-level node is a distribution that indicates the network certainty in different categories. In addition to recognizing input patterns, the HTM network can also propagate information down in the hierarchy using the belief propagation techniques that we described in earlier sections. Feeding information back in the hierarchy is used to segment the object from clutter and to locate the object in the image. More details about the training process for HTMs is available in [27] and in [8].


Towards a mathematical theory of cortical micro-circuits.

George D, Hawkins J - PLoS Comput. Biol. (2009)

Examples of training and testing images for an HTM network trained for visual object recognition.The top two rows are examples of training images. The bottom three rows are examples of correctly recognized test images. The last row shows test images that incorporated distracter backgrounds.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2749218&req=5

pcbi-1000532-g013: Examples of training and testing images for an HTM network trained for visual object recognition.The top two rows are examples of training images. The bottom three rows are examples of correctly recognized test images. The last row shows test images that incorporated distracter backgrounds.
Mentions: We used Numenta's NuPIC software environment to train a visual pattern recognition HTM network on which we tested the subjective contour effect. We started with an HTM network that was trained to recognize four different categories of objects: binoculars, cars, cell phones, and rubber ducks. This network had a three level HTM hierarchy. Figure 13 shows examples of training and testing images for these categories. When presented with a test image, the output from the top-level node is a distribution that indicates the network certainty in different categories. In addition to recognizing input patterns, the HTM network can also propagate information down in the hierarchy using the belief propagation techniques that we described in earlier sections. Feeding information back in the hierarchy is used to segment the object from clutter and to locate the object in the image. More details about the training process for HTMs is available in [27] and in [8].

Bottom Line: Anatomical data provide a contrasting set of organizational constraints.The combination of these two constraints suggests a theoretically derived interpretation for many anatomical and physiological features and predicts several others.We also discuss how the theory and the circuit can be extended to explain cortical features that are not explained by the current model and describe testable predictions that can be derived from the model.

View Article: PubMed Central - PubMed

Affiliation: Numenta Inc., Redwood City, California, United States of America. dgeorge@numenta.com

ABSTRACT
The theoretical setting of hierarchical Bayesian inference is gaining acceptance as a framework for understanding cortical computation. In this paper, we describe how Bayesian belief propagation in a spatio-temporal hierarchical model, called Hierarchical Temporal Memory (HTM), can lead to a mathematical model for cortical circuits. An HTM node is abstracted using a coincidence detector and a mixture of Markov chains. Bayesian belief propagation equations for such an HTM node define a set of functional constraints for a neuronal implementation. Anatomical data provide a contrasting set of organizational constraints. The combination of these two constraints suggests a theoretically derived interpretation for many anatomical and physiological features and predicts several others. We describe the pattern recognition capabilities of HTM networks and demonstrate the application of the derived circuits for modeling the subjective contour effect. We also discuss how the theory and the circuit can be extended to explain cortical features that are not explained by the current model and describe testable predictions that can be derived from the model.

Show MeSH
Related in: MedlinePlus