Data Elixir - Issue 457

State of AI Report 2023. Experiment interpretation. Digital afterlife. Multimodal systems. Research data management. Python nuances. ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌ ‌

ISSUE 457 · October 17, 2023   Trends State of AI Report 2023 This slide deck is one of the most comprehensive resources you'll find about the current state of AI. It distills what you need to know in AI research, commercial applications, politics, safety, and predictions for the coming year. It's dense but it's a very visual deck and easy to follow. Air Street Capital | Nathan Benaich   Sponsored Link New contrastive-learning methods for better data representation In a pair of papers, Amazon researchers show how to improve contrastive learning with decomposable loss functions and new geometric relationships in the representational space.       Reach Data Elixir readers by sponsoring an issue. Click here for details.   Posts & Tutorials Multimodality and Large Multimodal Models Language models have been all the rage in 2023 but being able to work with multimodal data (e.g. text, images, video, audio, etc) is essential in the real world. This is a great overview of multimodal systems, including what they are, typical tasks, training techniques, and active research. Chip Huyen   Experiment Interpretation and Extrapolation Great post about experiment interpretation and how to make decisions with the results. The common thread is that people are pretty good intuitive Bayesian reasoners, and so the best thing to do is to summarize the relevant evidence and let a human be the judge. Tom Cunningham   How well do you know Python? This interactive python quiz will test your Python knowledge and show you some nuances of the language along the way. jpglomot   How to help readers keep track of your color keys If you're presenting a lot of data and your visualization either scrolls or takes up a lot of space, readers may lose track of your color keys and misinterpret the data. This post explores ways to make sure that your color keys aren't forgotten. Datawrapper | Lisa Charlotte Muth   What Is Applied Intelligence? It’s harnessing the power of technology and human intelligence to make informed decisions that create a competitive advantage. It’s where data meets decision-making. At Georgetown, our Master’s in Applied Intelligence focuses on 4 key sectors: homeland security, cyber intelligence, law enforcement, and competitive business intelligence. Attend our webinar on Monday, October 23 at 12 p.m. ET and learn more. // sponsored   Papers A Survey of Large Language Models for Healthcare This is comprehensive overview of LLM applications in healthcare, starting with an exploration of the opportunities. Next, the paper surveys current applications, datasets, open source resources, methodologies, and the unique concerns of using LLMs in healthcare. arXiv | Kai He, Rui Mao et al.   Resources Data Management in Large-Scale Education Research This book by Crystal Lewis is a masterclass in managing data for large research projects. Its chapters follow the research life cycle and offer examples of best practices for each phase. Ultimately, following the ideas here should result in more secure, reproducible, and reliable data. Crystal Lewis   Geospatial Data Science with Julia This open-source book explores geospatial data science from the perspective of the Julia language. The book covers a variety of topics, starting with foundations of geospatial data and continues with transforms, interpolation, and practical applications. There are a lot of code examples too, which make it useful to see how Julia works.  Júlio Hoffimann   Outlier The Creepy New Digital Afterlife Industry "Digital afterlife" is a burgeoning new industry that uses data to bring back the dead. This article explores the landscape, the dangers, and the players. The possibilities here are as intriguing as they are alarming. IEEE Spectrum | Wendy H. Wong     How did you like this issue of Data Elixir? 👎    1     2     3     4     5   👍       Was this email forwarded to you?  Sign up here >>     Data Elixir, LLC P.O. Box 21255 Boulder, CO 80308 Data Elixir® is curated and maintained by Lon Riesberg. If you have questions or suggestions, just reply back to this email. Unsubscribe