Rebooting AI

What Hacker News says

1. mark_l_watsonAug 2023
   I have mixed feelings about Gary Marcus. I enjoyed reading Rebooting AI that he wrote with Ernest Davis, and he makes some valid points in this article.

   However, I think he ignores some practical use cases. I am not a good example since I am retired (I limit myself to about 6 to 8 hours a week consulting, and I am always working on writing a book) but I feel that I am at least 25% more efficient when using tools like GitHub Copilot integrated into Emacs and using Bing+ChatGPT or Bard for bouncing ideas off of, and sometimes for search. I find that if I carefully express an idea that I am thinking about that I usually get something at least interesting to think about and sometime very useful.

   I am excited by research on hybrid approaches to AGI, but also excited by what we have right now with OpenAI and Hugging Face APIs and models that can be self hosted.

   Show morethread
2. mark_l_watsonSep 2022
   Sure, he can sound strident but I still think Gary Marcus's riffing on the limitations of deep learning is important.

   The book "Rebooting AI" that he wrote with Ernest Davis is well worth reading if you are an AI practitioner (a term I use to describe myself). I think Marcus is also well worth following on Twitter to get a contrarian view (he re-tweeted me two weeks ago, so there is some overlap in our points of view).

   Way back when, I liked Roger Penrose's 1989 book "The Emperor's New Mind" even though some of the people I worked with thought he was a devil for writing that. I am much more optimistic than Marcus, but find his work useful and thoughtful.

   Show morethread
3. dalyMay 2022
   I read the GATO paper. I also saw the press hype. It is clear that the press has no idea about what GATO seems to be able to actually do.

   I can highly recommend the book "Rebooting AI" by Marcus and Davis (http://rebooting.ai/). They lay out a clear case why the current emphasis on things like deep learning won't get very far toward more general AI.

   In my research project the task was human/robot cooperation to change a car tire.

   Deep learning was useful for recognizing lugnuts and recognizing mis-threading due to torque issues. It is useful for things that are hard to describe such as how to ride a bicycle.

   GOFAI (good old fashioned AI) was useful for task modeling and planning such as setting the parking brake before attempting to work on the tire. It is useful for things that are easy to describe such as how to assemble a bicycle.

   Some tasks are learned by "compiling actions", an area without a lot of research as far as I know. If you want to learn to play a guitar the early stages are repetition. Gradually the process is "compiled into the hands". Typing is similar. Emacs is my ultimate example. Often when I try to tell someone about a good key sequence I have to act it out with my fingers before I can explain it because the sequence is "compiled into the fingers".

   People seem to think that Artificial General Intelligence is "in the near future". Musk, for example, is claiming that he will soon produce general purpose robots. He found out that self-driving was harder than he expected. Wait until he tries to make a robot enter a new house and make a cup of coffee. At best, both the coffee and the experience will be bitter.

   As to the question of robots being conscious, consider that people who drink a lot can "black out", still capable of doing many things, but being unaware that they have done them. "Doing" and "awareness" seem to be unrelated.

   The constant hype will eventually lead to a second "AI winter".

   Show morethread
4. mark_l_watsonApr 2022
   I like your comments, even though I fall into the DL fanboy camp.

   I hope that you read Gary Marcus’s and Ernest Davis’ book “Rebooting AI” - I think it would resonate with you, and it made me get a lot more interested in so-call “hybrid-AI” systems.

   thread
5. mark_l_watsonJun 2020
   This seems similar to the recent book Rebooting AI that he wrote with Ernest Davis.

   Good read (both paper and their book).

   A little off topic, but I have been an AI practitioner since the early 1980s. The beginning of my career was "symbolic" AI, the last 6 years were deep learning, and now I am also getting interested in hybrid AI which I mostly do in Common Lisp with some code calling out to Python and TensorFlow.

   thread
6. 2sk21Jan 2020
   Gary Marcus has articulated it much better than I could in his new book with Ernie Davis: https://www.amazon.com/Rebooting-AI-Building-Artificial-Inte...

   I really dislike the single minded focus on deep learning to the exclusion of everything else.

   thread