Book
?Download as PPTX, PDF?
1 like?539 views
AI is important for mobile devices because it can help users when they are traveling and away from other tools. While AI has made progress in games, question answering, and other domains, fully replicating human-level intelligence remains a major challenge. Estimates suggest parity between AI and human-level intelligence may be reached between 2030-2053 depending on factors like quantum computing advances, but the human brain is vastly more powerful than even the most advanced computers today. Continued progress will depend on overcoming limitations in computing power and algorithms.
Book
- 1. ARE THE ANDROIDS DREAMING YET? ARTIFICIAL INTELLIGENCE IN A MOBILE WORLD James Tagg Chief Technology Officer, Truphone
- 2. Im going to ask a three question. Why is AI so important to mobile? How close is AI to beating humans? Is it really going to happen? (feasibility)
- 3. First, a message from our sponsors Truphone
- 4. If you Travel Internationally
- 8. Back to the Book and AI
- 9. So why is AI so Important in Mobile? When we are stationary we can use tools. When we travel we need help.
- 10. How far has AI progressed? And how far does it need to go? Pose the question can a machine think creatively Or put another way Are humans WET computers?
- 11. Chess, Music, Watson and Siri ? Progress so far.. Deep Blue AARON Watson Emily Howell Jape
- 12. Human Connectome Project 0.5b Is this a copy? Analogy the Human Genome US project upload
- 13. Moores Law and the Human Brain
- 14. Classical Brain Parity Point 2047 RK 2053 JT SyNAPSE
- 15. Quantum Brain Hypothesis Photosynthesis is Quantum How about the Brain? Does the brain have the same structures? Where is the Light?
- 16. What about mobile (ARM) and mobile Moores Law is Irrelevant in Mobile. Mobile Chips Optimize for power. Run slightly behind Intel Desktop chips 100 Watts
- 17. POWER 2053 2020 2030 2040 2050 2060 2000 Estimate
- 18. POWER 2053 20202030204020502060 2080 Perspective we have a long way to go! Human brain is 85billion Neurons of analogy processing. Classical Parity Quantum Parity 100m 1b 10b 100b 1t 10t 100t 1p 10p 100p 1e 10e 100e What if Quantum is not enough? Oh and BTW we are already at 10Kw. The human brain is 20w.
- 22. Hilbert Problems 1900 10th Can all Diophantine Equations be solved? Can ALL mathematics be solved mechanically?
- 24. The Entscheidungsproblem Halting Problem has no solution
- 26. x + y = zn n n
- 27. 3 5 4
- 29. There is no algorithm which will find a solution to FLT if it is arbitrary to it. (by arbitrary we mean unknown) JT Matiyasevich Ruohonen and Baxa prove Exponentiated Diophantine equations can be rewritten as a regular Diophantine equations with the addition of an infinite set of terms therefore:-
- 30. FLT Proven The correct proof was published in May of 1995.
- 31. Paradox Wiles can not be a computer
- 32. Gracias Thank you james@taggs.com Free iBook (email me) Books 20, ?15, $20 English and Spanish!
- 33. Backup
- 34. How do we estimate the Classical to Quantum multiplier Only 1 Neuron but can: Swim. Hunt. Avoid Prey. Quess they need about the power of 1 remote control chip.
- 35. Definition of Arbitrary Mathematical Footing Fs=Fnc+S Computable evolution of a non-computable result
Editor's Notes
- #2: How brilliant are our minds?
- #6: Truphone SIM necklace
- #7: Quality when you travel
- #8: Convenience when you travel 1 bill 2 a local number (several if necessary)
- #12: So how good are computers
- #13: Fiber architecture of the brain. Measured from diffusion spectral imaging (DSI). The fibers are color-coded by direction: red = left-right, green = anterior-posterior, blue = through brain stem. www.humanconnectomeproject.org
- #15: Human brain Neuron firing rate
- #18: Extrapolate
- #19: And a bit more
- #20: In classical times people believed that we were only discovering things that had already been invented by the GODS. ? And when a Greek playwrite created a play they were merely documenting the human condition. ? They were DISCOVERING, NOT INVENTING.
- #21: But is AI a linear problem
- #23: The Maths
- #25: The unpronounsable paper
- #26: FERMAT Fermats last theorem is an exponential Diophantine equation. Everyone know Fermats last theorem?
- #27: Diophantine equations - everyone knows from their childhood. Why are these especially important? In mathematics, a?Diophantine equation?is a polynomial?equation. These ENCODE a wide range of problems, from word problems to more complicated problems in the sciences. They require creative thought. (go to triangle and hypercube)
- #30: Yuri Matiyasevich is best known for proving this C by solving Hilberts 10th problem with a negative answer. He proved that Diophantine Equations cannot be solved by machines.
- #31: After decades of work, Andrew Wiles DID solve Fermats last theorem. Here was the paradox C Humans CAN solve something a computer cannot.
- #33: In my book, Ya Suenan Los Androides?, Are the Androids Dreaming Yet? I discuss our incredible minds.