Intel

All Photos Twitter.com 14 hours ago
Heads? Tails? Or both at the same time? How wonderful! Learn about the power of qubits, the basic building blocks of #QuantumComputing. #BehindTheBrains
When safety meets scalability, we can help shift the autonomous driving revolution. All we need is an idea and Intel inside to do something wonderful.
Better explosions. Better battles. Better gaming. The new Intel 11th Gen #IntelCore processor with Iris Xe graphics is here. See how it’ll change the way you play @TencentGames.
By joining the @ChicagoQuantum Exchange, we’re committed to a wonderful collaboration with universities, laboratories, and more—to advance the science of quantum technologies.
With an idea and Intel inside, we can use AI data to forecast disruptions and envision solutions.
Intel promised a chip that delivered uncompromising performance on both CPU and GPU. It's clear they weren't kidding.​ extremetech.com/computing/3151…
Retweeted by Intel
What do you get when you mix quantum physics and chemistry? It looks like a steampunk chandelier, and it’s a wonderful new way to compute staggering amounts of data. #BehindTheBrains
We tested one of Intel’s Tiger Lake processors — and Xe Graphics are the real deal theverge.com/21440302/intel…
Retweeted by Intel
Most of today’s robots use only visual processing. Learn how our neuromorphic chips are bringing a sense of touch to robotics. #BehindTheBrains
Intel's 11th-Gen processor with Iris Xe Graphics is really that good bit.ly/2FtARwD
Retweeted by Intel
Mike Davies and the team at Intel Labs thank you for your time and questions. We hope you learned something new about neuromorphic computing today. Learn more about neuromorphic computing at Intel Labs: intel.ly/2FP94pT
We often have positions open for top algorithms, hardware, and software researchers. Search for “neuromorphic” at jobs.intel.com. —Mike Davies #BehindTheBrains
Q18. What opportunities are available at Intel for Neuromorphic Computing? #BehindTheBrains - @sunkissedbrownb
Our team is multidisciplinary with engineers, physicists, neuroscientists, and mathematicians collaborating closely with academics dedicated to the study of the brain. Our team is a fraction of the size of a traditional chip engineering team. —Mike Davies #BehindTheBrains
Q17. Can you talk about the team developing these neuromorphic chips? #BehindTheBrains
In neuromorphic networks, computation emerges from nonlinear dynamical interactions between millions of asynchronous neurons. This doesn’t fit the Von Neumann model at all and defies our intuitive understanding of computation. —Mike Davies #BehindTheBrains (2/2)
Yes! This is one of the biggest challenges facing the field. Not only do we need new languages, we need new programming models and abstractions because neuromorphic computation is so foreign compared to normal programming methods. —Mike Davies #BehindTheBrains (1/2)
Q16. Since the architecture is different, do you require new programming languages to make use of these chips? #BehindTheBrains (Ryan H., LinkedIn)
Over the long term, to advance computing to the next level, we’re going to need both technologies, not just one or the other. —Mike Davies #BehindTheBrains (3/3)
On the other hand, brains are terrible at solving numerically complex problems but excel at solving more approximate problems needing adaptation, generalization, associations, and learning. —Mike Davies #BehindTheBrains (2/3)
Our current computer architecture, dating back seven or eight decades, has proven extremely successful for solving highly precise and numeric kinds of problems, but less successful at more “brain-like” problems. —Mike Davies #BehindTheBrains (1/3)
Q15. Why is it critical that computing platforms function more like the human brain? #BehindTheBrains
Second, discovering the algorithms/applications that solve useful problems at that scale. —Mike Davies #BehindTheBrains (2/2)
From a technical standpoint, this is entirely feasible today. The main challenges are twofold. First, how to miniaturize such a system so it can be manufactured economically. —Mike Davies #BehindTheBrains (1/2)
Q14. How does Intel plan to scale into the billions of artificial neurons? #BehindTheBrains (Ryan H., LinkedIn)
Surgical assistance robots with emerging neuromorphic sensors and processing promise more precise control and better patient outcomes. —Mike Davies #BehindTheBrains (4/4)
Support for voice and visual processing that happens entirely inside your own computing device—without needing to send any personal data to the cloud. —Mike Davies #BehindTheBrains (3/4)
Personal computers that autonomously learn their user’s personal preferences, quirks, accents, vocabulary, and other unique traits. Things that can’t be programmed in the factory in a one-size-fits-all manner. —Mike Davies #BehindTheBrains (2/4)
There are many ways. For example, robots that move with grace, adaptation, and dexterity so they are more useful and safe around people – especially babies, the elderly, and the disabled. —Mike Davies #BehindTheBrains (1/4)
Q13. How will neuromorphic computing benefit regular people? #BehindTheBrains
Loihi solves problems in dramatically different ways from conventional computers. On many examples it’s showing orders of magnitude gains in energy and speed, which could be a game changer. It’s a new paradigm that’s only beginning to be explored. —Mike Davies #BehindTheBrains
Q12. How is neuromorphic computing going to surpass the limitations of traditional computing? #BehindTheBrains
This will provide many opportunities for a B2B marketplace, mainly in the domain of software, modeling, and system integration. —Mike Davies #BehindTheBrains (2/2)
A complete ecosystem with an active developer community will be necessary to move neuromorphic computing into mainstream commercial applications. —Mike Davies #BehindTheBrains (1/2)
Q11. What does the future hold for the neuromorphic computing B2B marketplace? #BehindTheBrains - @SakirudeenOlal2
Pohoiki Springs is a neuromorphic research system that integrates 768 Loihi processors onto a chassis the size of five servers. The system reaches the equivalent of 100M neurons and provides the scale needed for advanced AI research. —Mike Davies #BehindTheBrains
Q10. What is Pohoiki Springs? #BehindTheBrains
Loihi is the fourth chip in Intel’s neuromorphic program and contains the equivalent of 130,000 neurons. We’ve found many examples where Loihi outperforms conventional CPUs and specialized AI chips byvery large factors. —Mike Davies #BehindTheBrains
Q9. Loihi is the latest neuromorphic chip from Intel. How is this chip being used today? #BehindTheBrains
Regarding odor perception, Professor Thomas Cleland at Cornell University and Germany’s Fraunhofer Institute are most actively pursuing applications, mainly for explosives and other hazardous chemical detection purposes. —Mike Davies #BehindTheBrains (2/2)
Applications for all manner of sensor modalities are being explored, including vision, audio, smell, tactile, EEG/EMG, and wireless RF. —Mike Davies #BehindTheBrains (1/2)
Q8. I read the Loihi chips were trained to discern smell. Are there plans for other sensory tasks and does Intel see this being used in narcotics/drug detection? #BehindTheBrains (-Ryan H. LinkedIn)
We consider the mathematical models and principles that neuroscientists have developed describing brain operation and adapt them to the circuits of silicon transistors that we can build with today’s chip technology. – Mike Davies #BehindTheBrains
Q7. How do you use our understanding of the brain to create a neuromorphic chip? #BehindTheBrains
Neuromorphic computing, as a field, mostly refers to a new type of hardware architecture modeled after the brain. How this new type of computer architecture is programmed will determine whether and to what degree its behavior is biased. —Mike Davies #BehindTheBrains
Q6. Current implementation of AIs already has bias, will neuromorphic computing further escalate that bias? #BehindTheBrains - @vasanth59447050
It is possible to run traditional Artificial Neural Networks on neuromorphic chips like Loihi, but the gains are modest (if nonexistent) since those networks were developed for conventional computing architectures and synchronous data formats. —Mike Davies #BehindTheBrains (3/3)
For example, networks that solve optimization and constraint satisfaction problems — like Sudoku. —Mike Davies #BehindTheBrains (2/3)
We’ve measured huge gains – up to 100x faster and 1000x lower energy – but only with new kinds of neural networks that go beyond traditional Artificial Neural Networks. —Mike Davies #BehindTheBrains (1/3)
Twiends™ uses the Twitter™ API, displays it's logo & trademarks, and is not endorsed or certified by them. These items remain the property of Twitter. We do not sell followers, we only provide display advertising. Bots & fake accounts are not permitted on twiends. © 2009
Grow Your Twitter Free
Want To Grow Your Twitter?
We help other people find and follow you on Twitter.
Key Info:
Started in 2009
Over 6 million signups
Country targeting provided
We never auto tweet to your timeline
We never auto follow others
We actively moderate our community
Please Share
Please upgrade your browser  chrome