https://lilianweng.github.io/posts/2022-04-15-data-gen/ Learning with not Enough Data Part 3: Data GenerationApril 15, 2022Lilian Weng “Here comes the Part 3 on learning with not enough data (Previous: Part 1 and Part 2).
What I Read: Understanding, Simple AI
https://www.quantamagazine.org/researchers-glimpse-how-ai-gets-so-good-at-language-processing-20220414/ Researchers Gain New Understanding From Simple AIMordechai RorvigStaff WriterApril 14, 2022 “Language processing programs are notoriously hard to interpret, but smaller versions can provide important insights into how they
What I Read: forecasting, quantile functions
https://www.amazon.science/blog/improving-forecasting-by-learning-quantile-functions Improving forecasting by learning quantile functionsBy Youngsuk Park, François-Xavier AubetMarch 31, 2022 “Learning the complete quantile function, which maps probabilities to variable values, rather than building separate models for
What I Read: Graph ML, missing node features
https://blog.twitter.com/engineering/en_us/topics/insights/2022/graph-machine-learning-with-missing-node-features Graph machine learning with missing node featuresEmanuele RossiMaria GorinovaThursday, 17 March 2022 “…GNNs typically run under the assumption of a full set of features available for all nodes. In
What I Read: Policy Regulariser, Adversary
https://deepmindsafetyresearch.medium.com/your-policy-regulariser-is-secretly-an-adversary-14684c743d45 Your Policy Regulariser is Secretly an AdversaryDeepMind Safety ResearchMar 24 By Rob Brekelmans, Tim Genewein, Jordi Grau-Moya, Grégoire Delétang, Markus Kunesch, Shane Legg, Pedro A. Ortega“Policy regularisation can be