AI - AI Dynamics

→ View original post on X — @ronald_vanloon

Google and Microsoft Launch Competing Generative AI Products

By

@ronald_vanloon

–

10 March 2023 7h50

Google and Microsoft prepare dueling generative #AI debuts
by @sharongoldman @VentureBeat Go to: https://
buff.ly/3wXJ9Uz #BigData #ArtificialIntelligence #Digital #DataScience cc: @ronald_vanloon @pbalakrishnarao @karpathy

→ View original post on X — @ronald_vanloon

Amazing Self-Driving Bicycle Innovation in AI and Machine Learning

By

@ronald_vanloon

–

10 March 2023 7h27

Amazing #SelfDriving bicycle
via @FrRonconi #AI #MachineLearning #ArtificialIntelligence #ML #MI #DL #Innovation #Automotive #Transport #FutureofWork

cc: @ronald_vanloon @pbalakrishnarao @levie pic.twitter.com/BEtoU3bAHT
— Ronald van Loon (@Ronald_vanLoon) 10 mars 2023

Amazing #SelfDriving bicycle
via @FrRonconi #AI #MachineLearning #ArtificialIntelligence #ML #MI #DL #Innovation #Automotive #Transport #FutureofWork cc: @ronald_vanloon @pbalakrishnarao @levie

Accurate CO2e Measurements for ML Models Matter

By

–

10 March 2023 7h08

An important thread about misrepresentation of CO2e numbers for different ML models. I believe it's good to use actual measurements of this data, rather than estimates that are flawed in a variety of ways.

Google Cloud Infrastructure and AI Hardware Energy Efficiency Comparison

By

–

10 March 2023 7h00

All the infrastructure is accessible to external entities through our Google Cloud Platform. Table 1 in https://
arxiv.org/abs/2104.10350 has columns that enables you to compare moving from avg US datacenter/electrical grid to Google datacenter, & also moving from P100 to TPUv2.

→ View original post on X — @officialindiaai

India’s AI Leadership: Key Organizations Driving Digital Transformation

By

@officialindiaai

–

10 March 2023 6h45

@ZuruServices @weareshAIp @iMeritDigital @TikaData @_DigitalIndia @GoI_MeitY @nasscom @NeGD_GoI

→ View original post on X — @officialindiaai

Five Indian Startups Leading Data Labelling for AI Models

By

@officialindiaai

–

10 March 2023 6h45

An essential step in creating a successful ML model is labelling the data. Data labelling is the backbone of AI, and these five Indian startups are paving the way for the future. Check out what they have to offer here: https://
bit.ly/3l2Vnc5 @jeevprabnivash @taskmonk

Google’s Flawed Training Cost Estimates Criticized by Researcher

By

–

10 March 2023 6h45

I have raised this exact point in the past with some Google authors, saying that it wasn't reasonable to use flawed estimates that were off by a factor of 3261X from the actual cost of training this language model, and that they should fix this before publishing. Sigh.

Language Model Training Efficiency on Google TPU Hardware

By

–

10 March 2023 6h37

… and off by ~118,000X if you consider training that language model on a Google TPUv2 in our Iowa datacenter. See Table 1 and Appendix C and D of https://
arxiv.org/abs/2104.10350 for more details.

Language Model Training Emissions 3261X Lower Than Previous Estimates

By

–

10 March 2023 6h32

The actual emissions for training a language model of the scale examined by Strubell et al. is 3261X smaller than this flawed estimate even if you consider the "P100 in average US data center on average electrical grid" scenario they were evaluating
…