Summary of Explainable Moral Values: a Neuro-symbolic Approach to Value Classification, by Nicolas Lazzari et al.
Explainable Moral Values: a neuro-symbolic approach to value classification
by Nicolas Lazzari, Stefano De Giorgis, Aldo Gangemi, Valentina Presutti
First submitted to arxiv on: 16 Oct 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG)
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 This paper combines ontology-based reasoning and Machine Learning techniques to develop explainable value classification. It uses the Moral Foundations Theory’s ontological formalization of moral values, along with the DnS Ontology Design Pattern, to infer values that are satisfied by a given sentence using the Sandra neuro-symbolic reasoner. The system automatically generates sentences and their structured representations using an open-source Large Language Model. The inferred descriptions are used for value detection in sentences. The results show that relying solely on the reasoner’s inference produces comparable explainable classification to other approaches. Combining the reasoner’s inferences with distributional semantics methods outperforms all baselines, including neural network-based models. A visualization tool is built to explore theory-based values classification and is publicly available. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses special computer reasoning and machine learning techniques to understand what things mean and why they are important. It looks at how we think about good and bad actions by using a special framework called Moral Foundations Theory. The system tries to figure out which moral value (like fairness or loyalty) is important for each sentence it looks at. To do this, it uses a big language model to generate sentences and then checks if certain values are important based on those sentences. This helps make sense of why we think something is good or bad. |
Keywords
» Artificial intelligence » Classification » Inference » Language model » Large language model » Machine learning » Neural network » Semantics
