Summary of Implementation and Evaluation Of a Gradient Descent-trained Defensible Blackboard Architecture System, by Jordan Milbrath et al.
Implementation and Evaluation of a Gradient Descent-Trained Defensible Blackboard Architecture System
by Jordan Milbrath, Jonathan Rivard, Jeremy Straub
First submitted to arxiv on: 17 Apr 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A combined implementation that leverages the strengths of rule-fact expert systems and neural networks is proposed, utilizing gradient descent to train a rule-fact expert system. This approach combines the benefits of human domain expertise with the ability to learn from presented data. The Blackboard Architecture, which adds actualization capabilities to expert systems, is also integrated. Additionally, activation functions are introduced for defensible artificial intelligence systems and a new best path-based training algorithm is implemented and evaluated. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a way to make AI more reliable and understandable by combining different techniques. It uses something called gradient descent to train a rule-fact expert system, which helps it learn from data. The Blackboard Architecture is also used, which allows the system to actually use its knowledge. To make the system even better, new ways of training are introduced, like using activation functions. |
Keywords
* Artificial intelligence * Gradient descent
