Neuro-Symbolic AI - combining Deep Learning with Symbolic Logic for common-sense understanding
Dr Maria Aretoulaki
Explainable Responsible GenAI & Conversational AI, AI Policy @GlobalLogic/HITACHI, 30yrs in Chatbot Design, Language & Prompt Engineering (LangOps), AI Person of the Year 2023 Finalist, Conversational AI Leader 2023
I recently read an article in MIT Technology Review entitled "A hybrid AI model lets it reason about the world’s physics like a child". In it, a so-called "Neuro-Symbolic model" is presented, a new AI approach that combines both Deep Learning and Symbolic Logic.
"it uses a neural network to recognize the colors, shapes, and materials of the objects and a symbolic system to understand the physics of their movements and the causal relationships between them. It outperformed existing models across all categories of questions."
The motivation behind this approach was two-fold:
- Deep Learning is superb at pattern matching, but bad at actually understanding the data it processes.
- Symbolic Logic can capture and model really well human reasoning, correlations and interdependencies between actions, events, objects, people, and "common sense" or even objective rigorous Physics understanding.
Thus, the natural next step is to combine both and leverage the strengths of each in a Hybrid model:
"Deep Learning excels at scalability and pattern recognition; symbolic systems are better at abstraction and reasoning."
Funnily enough that was the approach I took myself for my PhD research back in 1993-96, when I conceived of a Hybrid architecture for automatic Text Summarisation that uses Artificial Neural Networks (ANNs - which have more recently turned into Deep Learning) and Symbolic Reasoning. It was a very ambitious undertaking, but it seemed to be to be the next natural step in AI methodologies. The title of my PhD was "COSY-MATS: A Hybrid Connectionist-Symbolic Approach To The Pragmatic Analysis Of Texts For Their Automatic Summarisation" (University of Manchester, UK, 1996) and you can find the whole Thesis here.
#NLP #NLU #Summarisation #Summarization #AI #ML #MachineLearning #DeepLearning #SymbolicAI #Logic #Hybrid #Connectionist #NeuroSymbolic #AImodels #CosyMats #ConnectionistSymbolic #ANNs #NeuralNetworks #Manchester
Expert in Residence Imperial Enterprise Lab -- Imperial College London
4 年I have been interested in this hybrid approach as well, particularly for live chat automation. But the difficulty of keeping the Symbolic portion up to date is still with us. My thought is to use the Symbolic portion purely for adding new business rules, ad hoc, until the connectionist data is sufficient for retraining to incorporate those new rules implicitly.
Chief Strategy and Research Officer at Talkmap
4 年It’s interesting to see symbolic approaches seing back into vogue as a potential way to fill the gap for statistical models. It makes sense, though it seems like a tricky proposition that will take a lot of experimentation to find a good approach. Thanks for writing a summary of this article.
Helping senior living communities leverage cutting edge technology safely and cost effectively.
4 年Thank you for summarizing and sharing. We should all do more of this.