““You can’t get to the moon by climbing successively taller trees.””
Artificial Intelligence is a very hot topic. In my opinion there are two polarized camps on the current state of AI: those who think that we are on the brisk of creating a superhuman systems within next couple of years and those who think that we are nowhere close to a system that can be labeled intelligent. The demonstration of this duopoly can be the notorious debate between Yann LeCunn (chief AI scientist of Facebook) and Gary Marcus(scientist and professor at NYU) . The debate demonstrates completely different approaches about what AI is from tools and techniques to general philosophy. I am a big fan of Gary Marcus perspective and have recently overshared his critical piece on GPT-2. This week I have finished the latest book by professor Marcus: Rebooting AI: Building Artificial Intelligence We Can Trust which is a very simple yet critical summarization without the hype and over-exageration.
“The scenario that the robots will become super intelligent and enslave humans makes about as much sense as the worry that since jet planes have surpassed the flying ability of eagles, someday they will swoop the sky and seize our cattle”
Here is the list of suggested readings about AI, robotics and common sense from Gary Marcus (a lot of the books are free and accessible online):
- General on AI: Artificial Intelligence: A Modern Approach (3rd Edition): Stuart Russell, Peter Norvig: 8601419506989: Amazon.com: Books
- Scepticism about AI: /Computer Power and Human Reason/ Computer Power and Human Reason - Wikipedia
- What Computers Can’t Do Book by Hubert Dreyfus https://www.amazon.com/What-Computers-Still-Cant-Artificial/dp/0262540673
- AI systems that read: Speech and Language Processing Jurafsky: Speech and Language Processing
- Introduction to Information Retrieval: Introduction to Information Retrieval
- Robotics: Towards Robotic Manipulation by Natthew Mason: http://www.cs.cmu.edu/afs/cs/academic/class/16741-s07/www/Mason2018.pdf
- Robotics: Modern Robotics bt Kevin Lyrch: http://hades.mech.northwestern.edu/images/7/7f/MR.pdf
- Robotics: Planning Algorithms Stephen La Valle Planning Algorithms / Motion Planning
- On how Mind works: The Language Instinct Steven Pincker: http://f.javier.io/rep/books/The-Language-Instinct-How-the-Mind-Creates-Language,-Steven-Pinker.pdf
- More on how mind works: human knowledge its scope and limits bertrand russell https://www.amazon.com/Human-Knowledge-Limits-Routledge-Classics/dp/0415474442
- On common sense: Common Sense, the Turing Test, and the Quest for Real AI | The MIT Press by Hector J. Lebesque
- More on Common Sense : Handbook of Knowledge Representation https://dai.fmph.uniba.sk/~sefranek/kri/handbook/handbook_of_kr.pdf Frank Harmelen
My backlog reading list has exploded again.
I don’t know how I made it through business school without knowing who was John Boyd. Arguably “greatest strategist since Sun Tzu”, he has invented “OODA” (Observe, Orient, Decide, and Act) Loop. Though very intuitive and “obvious” from the first glance, the “Orient” box looks particularly interesting and deliberate. In this phase, Boyd proposed, happens information filtering through culture, genetics and previous experience.
Side note: here is a good podcast episode on who was John Boyd