Demis Hassabis: Towards General Artificial Intelligence – talk at Center for Brains, Minds and Machines (CBMM). [Background: r. Demis Hassabis is the Co-Founder and CEO of DeepMind, the world’s leading General Artificial Intelligence (AI) company, which was acquired by Google in 2014 in their largest ever European acquisition.
The talk draws on Demis’ eclectic experiences as an AI researcher, neuroscientist and video games designer.
the 1 1/2 movie about AlphaGo, Deep Mind’s AI program that beat the best GO player in the world. AMAZING . MUST WATCH if you are really into real AI.
If you want to really understand what happened and how, and how Deepmind really works … you’ve got to watch the documentary.
Lots of information about the game of GO … a great way to bring 8000 years of GO history and the future together
Kai-Fu Lee TED Talk about Chinese AI, here. Worth watching. Many insights!
WATCH Demis Hassabis: creativity and AI TALK – The Rothschild Foundation Lecture
DEEP MIND and DEEP FUTURE
according to popular legends and urban myths … Amos Tversky is said to have said …
My colleagues, they study artificial intelligence; me, I study natural stupidity.
this, from CoEvolving Innovations which seems like a fascinating resource.
The blog entry there talks about Daniel Kahneman and Amos Tversky.
The topic is fascinating. The question of how intelligence and stupidity are related is fascinating.
There’s also a reference to Daniel Kahneman, Paul Slovic, and Amos Tversky book Judgment under uncertainty: Heuristics and biases, that I now feel compelled to investigate
interesting factoid …Kahneman was awarded the 2002 Nobel Prize in economic sciences despite being a psychologist, not an economist. Which goes to show you … that Forrest Gump’s Mom was right “Life is like a box of chocolates. You never know what you’re gonna get.”
just announce on the Google AI Blog …/
this is pretty exciting …hope to this grow and have fruitful implementation on the Google search engine.
this is what the Google AI researchers are saying
…. there are currently no large, publicly available sources of naturally occurring questions (i.e. questions asked by a person seeking information) and answers that can be used to train and evaluate QA models. This is because assembling a high-quality dataset for question answering requires a large source of real questions and significant human effort in finding correct answers.
To help spur research advances in QA, we are excited to announce Natural Questions (NQ), a new, large-scale corpus for training and evaluating open-domain question answering systems, and the first to replicate the end-to-end process in which people find answers to questions. NQ is large, consisting of 300,000 naturally occurring questions, along with human annotated answers from Wikipedia pages, to be used in training QA systems. We have additionally included 16,000 examples where answers (to the same questions) are provided by 5 different annotators,
I am really looking forward to digging into this …good questions and good answers are definitely part of the key for solving some great puzzles ….
have fun …