Field Note #151. On "Neurosymbolic AI"

Source: Artur d'Avila Garcez and Luis Lamb, "Neurosybmolic AI: The 3rd Wave" avail. at online at arXiv (Dec. 2020)

The Context.

  1. To build human-level intelligence system, we need AI systems with a reasoning layer along with statistical ML, model-blind deep learning. (See, e.g. Field Note #150)
  2. Rich AI system should be semantically sound, explainable, and trustworthy
  3. Current AI paradigm skews heavily toward statistical ML, but "one would be misguided to conflate all of machine learning or dismiss the role of symbolic logic, which remains the most powerful and adequate representation for the analysis of computational systems." (p. 14, emphasis added)
  4. Neurosymbolic AI has the potential to unify learning and reasoning

The Map.

You've successfully subscribed to Sandeep C. Ramesh
Great! Next, complete checkout for full access to Sandeep C. Ramesh
Welcome back! You've successfully signed in.
Unable to sign you in. Please try again.
Success! Your account is fully activated, you now have access to all content.
Error! Stripe checkout failed.
Success! Your billing info is updated.
Error! Billing info update failed.