https://towardsdatascience.com/can-a-neural-network-train-other-networks-cf371be516c6?gi=4765ee11be44 Can a neural network train other networks?An introduction to knowledge distillationTivadar DankaOct 5 “Now you have a huge model, which, although performs excellently, there is no way to deploy
What I Read: Transformers for Image Recognition
https://medium.com/swlh/an-image-is-worth-16×16-words-transformers-for-image-recognition-at-scale-brief-review-of-the-8770a636c6a8 An Image Is Worth 16×16 Words: Transformers for Image Recognition at Scale (Brief Review of the ICLR 2021 Paper)Stan KriventsovOct 9 “The reason attention models haven’t been doing better
What I Read: automatic differentiation with graphs
https://ai.facebook.com/blog/a-new-open-source-framework-for-automatic-differentiation-with-graphs A new open source framework for automatic differentiation with graphsOctober 8th, 2020 “Just as PyTorch provides a framework for automatic differentiation with tensors, GTN provides such a framework for
What I Read: Production with Deep Semi-Supervised Learning
https://towardsdatascience.com/from-research-to-production-with-deep-semi-supervised-learning-7caaedc39093 From Research to Production with Deep Semi-Supervised LearningVarun NairSep 25 “Semi-supervised learning (SSL), a subfield that combines both supervised and unsupervised learning, has grown in popularity in the deep
What I Read: AI’s limitations
https://www.economist.com/technology-quarterly/2020/06/11/an-understanding-of-ais-limitations-is-starting-to-sink-in Technology QuarterlyJun 11th 2020 editionAn understanding of AI’s limitations is starting to sink inAfter years of hype, many people feel AI has failed to deliver, says Tim Cross “Surveying
What I Read: This AI learns by reading the web
https://www.technologyreview.com/2020/09/04/1008156/knowledge-graph-ai-reads-web-machine-learning-natural-language-processing/ This know-it-all AI learns by reading the entire web nonstopDiffbot is building the biggest-ever knowledge graph by applying image recognition and natural-language processing to billions of web pages.Will Douglas
What I Read: Best Practices for Building Machine Learning at Scale
https://medium.com/dataseries/linkedins-pro-ml-architecture-summarizes-best-practices-for-building-machine-learning-at-scale-77fcb6afc9ec LinkedIn’s Pro-ML Architecture Summarizes Best Practices for Building Machine Learning at ScaleThe reference architecture is powering mission critical machine learning workflows within LinkedIn.Jesus RodriguezSep 17 “Building machine learning solutions