Skip to content
Zhengyuan Zhu
Go back

HTM_theory

hierarchical structure.(need a pic)and each region is performing the same set of processes on the input data.

[SDR] sparse distributed representations(0 or 1)

Input: 1. Motor commands 2. sensory input

Encoder: takes a datatype and converts it into a sparse distributed representations.

Temporal means that systems learn continuously, every time it receives input it is attemptiing to predict what is going to happen next.

Capacity=n!/w!(n-w)!

OVERLAP/UNION Similarity can be represented by overlap/union? of SDR.

MATCH

Scalar Encoder(retina/cochlea)

Data-time encoder


Share this post on:

Previous Post
Video Captioning Based on Hierarchical Reinforcement Learning
Next Post
Video Captioning Based on Temporal Structure
Jack the orange tabby cat
I'm Jack ๐Ÿงก
Luna the tuxedo cat
I'm Luna! ๐Ÿ–ค