Chapter
Interoperability between Deep Neural Net Frameworks and Open-Cog Hypergraph
2:01:40 - 2:11:10 (09:29)
The integration between deep neural net frameworks and open-cog hypergraph was clunky due to the limitations of the outdated Torch computation graph software. In addition, the system wasn't designed for modern scalability but it is still possible to have types as variables and do type checking among complex higher-order types, albeit, it's a slow process.
Clips
The challenge is to represent the different types of knowledge needed for various learning, such as declarative, procedural, and attentional, in a way that allows for cross-learning to occur between different types of memory.
2:01:40 - 2:04:35 (02:54)
Summary
The challenge is to represent the different types of knowledge needed for various learning, such as declarative, procedural, and attentional, in a way that allows for cross-learning to occur between different types of memory. This involves finding a way to make representations of maps in the memory matrix work together with representations needed for visual pattern recognition in neural networks.
ChapterInteroperability between Deep Neural Net Frameworks and Open-Cog Hypergraph
Episode#103 – Ben Goertzel: Artificial General Intelligence
PodcastLex Fridman Podcast
The speaker discusses rebuilding the existing AI system, which they previously used in their consulting business, to create true Artificial General Intelligence (AGI) that can mimic human-like intelligence using the OpenCog architecture while incorporating logic as a key part of the system.
2:04:35 - 2:05:35 (01:00)
Summary
The speaker discusses rebuilding the existing AI system, which they previously used in their consulting business, to create true Artificial General Intelligence (AGI) that can mimic human-like intelligence using the OpenCog architecture while incorporating logic as a key part of the system.
ChapterInteroperability between Deep Neural Net Frameworks and Open-Cog Hypergraph
Episode#103 – Ben Goertzel: Artificial General Intelligence
PodcastLex Fridman Podcast
The combination of deep neural models and symbolic systems required the team to feed the outputs of deep neural models into an open-cog symbolic representation, then perform pattern mining and reasoning on the different camera feeds, using a project for Cisco that watched street scenes as an example.
2:05:35 - 2:07:44 (02:09)
Summary
The combination of deep neural models and symbolic systems required the team to feed the outputs of deep neural models into an open-cog symbolic representation, then perform pattern mining and reasoning on the different camera feeds, using a project for Cisco that watched street scenes as an example. Real-time access was needed for accessing computation graphs on Torch in a hypergraph.
ChapterInteroperability between Deep Neural Net Frameworks and Open-Cog Hypergraph
Episode#103 – Ben Goertzel: Artificial General Intelligence
PodcastLex Fridman Podcast
The speaker discusses the challenges of achieving efficient interoperation between the computation graphs of deep neural net frameworks and open-cog hypergraph, as well as the need for effective scalability in open-cog hypergraph for distributed computing.
2:07:44 - 2:11:10 (03:26)
Summary
The speaker discusses the challenges of achieving efficient interoperation between the computation graphs of deep neural net frameworks and open-cog hypergraph, as well as the need for effective scalability in open-cog hypergraph for distributed computing.