It redesigned deep-learning algorithms to run more efficiently on a CPU by
utilizing the chips’ large available memory and complex cores. While the approach loses the speed
achieved through a GPU’s parallelization, it reportedly gains back about the same amount of time by
eliminating the need to ferry data on and off the chip. The algorithms can run on CPUs “at GPU
speeds,” the company says—but at a fraction of the cost.
## commercial companies
- [Intel dumps its Nervana neural network processors for Habana’s AI chips](https://siliconangle.com/2020/02/02/intel-dumps-nervana-neural-network-processors-habanas-ai-chips/)
[The First AI Chip Startup with Naveen Rao, Nervana Systems - FYI - For Your Innovation](https://podcasts.apple.com/gb/podcast/fyi-for-your-innovation/id1271691895?i=1000499936500)
- 2020 [**Apple buys edge-based AI startup Xnor.ai for a reported 200M**](https://techcrunch.com/2020/01/15/apple-buys-edge-based-ai-startup-xnor-ai-for-a-reported-200m/?guccounter=1&guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAACQ2Pdxdyw0ekFeJaOYj2NJxsmksYYBVNHe0aMaWbOAPU0OzwbdW_NEKsFqvKr10qKgbpo5JiFkIZqRfwQ_3PSfbIPCYi4Oh_XcCslqeWKtMEde65htpixUV_iUQdHbCVzMGShDwMBOnjGsPAlZl2GL8fsEP4ZKpITAlS136B07-)
https://www.crunchbase.com/organization/xnor-ai
2020 [Exclusive: Apple acquires Xnor.ai, edge AI spin-out from Paul Allen’s AI2, for price in 200M range](https://www.geekwire.com/2020/exclusive-apple-acquires-xnor-ai-edge-ai-spin-paul-allens-ai2-price-200m-range/)
[Xnor’s saltine-sized, solar-powered AI hardware redefines the edge](https://techcrunch.com/2019/02/13/xnors-saltine-sized-solar-powered-ai-hardware-redefines-the-edge/?guccounter=1)
- https://groq.com/about/
> ...Google’s TPU effort as a 20% project where he designed and implemented the core elements of the original chip
- https://www.untether.ai/
In-memory computing with Flash.... 100m, but results???
https://en.wikipedia.org/wiki/Digital_signal_processor#Overview
- https://plumerai.com/
**BNN with microcontrollers**
podcast [Using Existing Edge Hardware for New AI Capabilities - with Roeland Nusselder of Plumerai](https://podcasts.apple.com/gb/podcast/ai-in-business/id670771965?i=1000492081227)
- https://landing.ai/
- NeuralMagic [The startup making deep learning possible without specialized hardware](https://s3.amazonaws.com/knowen-production/big_attachments/bde6e89c7c2f1a0a1b4e8fbc30d403ae/Untitled+42.pdf)
> It redesigned deep-learning algorithms to run more efficiently on a CPU by
> utilizing the chips’ large available memory and complex cores. While the approach loses the speed
> achieved through a GPU’s parallelization, it reportedly gains back about the same amount of time by
> eliminating the need to ferry data on and off the chip. The algorithms can run on CPUs “at GPU
> speeds,” the company says—but at a fraction of the cost.
- https://www.mythic-ai.com/, see in [more detail here](https://knowen.org/nodes/27246)
## NY Times articles
[Moore’s Law Running Out of Room, Tech Looks for a Successor](https://knowen-production.s3.amazonaws.com/uploads/attachment/file/5063/Moores%2BLaw%2BRunning%2BOut%2Bof%2BRoom%2BTech%2BLooks%2Bfor%2Ba%2BSuccessor%2B-%2BThe%2BNew%2BYork%2BTimes.pdf)
[Chips Off the Old Block: Computers Are Taking Design Cues From Human Brains](https://knowen-production.s3.amazonaws.com/uploads/attachment/file/5064/Chips%2BOff%2Bthe%2BOld%2BBlock%2BComputers%2BAre%2BTaking%2BDesign%2BCues%2BFrom%2BHuman%2BBrains%2B-%2BThe%2BNew%2BYork%2BTimes.pdf)
## conferences
https://aihardwaresummit.com/events/ai-hardware-summit-2021#prospectus
https://www.aihardwaresummit.com/events/edge-ai-summit-2021
https://www.kisacoresearch.com/sites/default/files/agendas/kr_tech_sponsorship_prospectus_-_public.pdf
http://isscc.org/