Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2404.11714 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910414401437696 |
|---|---|
| author | Milbrath, Jordan Rivard, Jonathan Straub, Jeremy |
| author_facet | Milbrath, Jordan Rivard, Jonathan Straub, Jeremy |
| contents | A variety of forms of artificial intelligence systems have been developed. Two well-known techniques are neural networks and rule-fact expert systems. The former can be trained from presented data while the latter is typically developed by human domain experts. A combined implementation that uses gradient descent to train a rule-fact expert system has been previously proposed. A related system type, the Blackboard Architecture, adds an actualization capability to expert systems. This paper proposes and evaluates the incorporation of a defensible-style gradient descent training capability into the Blackboard Architecture. It also introduces the use of activation functions for defensible artificial intelligence systems and implements and evaluates a new best path-based training algorithm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_11714 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Implementation and Evaluation of a Gradient Descent-Trained Defensible Blackboard Architecture System Milbrath, Jordan Rivard, Jonathan Straub, Jeremy Artificial Intelligence A variety of forms of artificial intelligence systems have been developed. Two well-known techniques are neural networks and rule-fact expert systems. The former can be trained from presented data while the latter is typically developed by human domain experts. A combined implementation that uses gradient descent to train a rule-fact expert system has been previously proposed. A related system type, the Blackboard Architecture, adds an actualization capability to expert systems. This paper proposes and evaluates the incorporation of a defensible-style gradient descent training capability into the Blackboard Architecture. It also introduces the use of activation functions for defensible artificial intelligence systems and implements and evaluates a new best path-based training algorithm. |
| title | Implementation and Evaluation of a Gradient Descent-Trained Defensible Blackboard Architecture System |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2404.11714 |