Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Optimize inference and learning process using mirror connections. #10

Open
marty1885 opened this issue May 24, 2019 · 0 comments
Open
Labels
need discussion need some discussion optimization make stuff faster! Spatial Pooling Regarding Spatial Pooling Temporal Memory Regarding Temporal Memory

Comments

@marty1885
Copy link
Member

In the Bachelor project by Ali Kaan Sungar, Hierarchical Temporal Memory Software Agent section 5.4.3. He purposes that mirrored synapses (maintaining 2 sets of identical synapses but running in the opposite direction) are a major optimization. Allowing a vastly reduced computation requirement.

The same tech can be also used in Etaler to enhance the performance further. But maintaining 2 lists of synapses is complicated under Etaler's DOD/functional API.

@marty1885 marty1885 added need discussion need some discussion optimization make stuff faster! labels May 24, 2019
@marty1885 marty1885 added Spatial Pooling Regarding Spatial Pooling Temporal Memory Regarding Temporal Memory labels Aug 13, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
need discussion need some discussion optimization make stuff faster! Spatial Pooling Regarding Spatial Pooling Temporal Memory Regarding Temporal Memory
Projects
None yet
Development

No branches or pull requests

1 participant