Emergent Methods’ Post

Data leaks are common when hackathons are teaching people to create them. They are also common when AIs are using neurodusts to hack and automate mind and brain data. Hackathons teach people to hack more, better, faster.

Together with my teammates Mesut Duman and Burhan Yıldız, we completed the Presentation stage of the Datathon-2024 competition as the We Bears team, securing 4th place. 🎉 The challenging competition, with 575 participants in 364 teams, was organized by BTK Akademi in collaboration with Google and the Türkiye Entrepreneurship Foundation. Achieving 4th place in this tough contest was an important step in applying what we’ve learned and solidifying our expertise in data science. You can check our presentation and codes from the links below; Codes; https://lnkd.in/dDcxkVYf Presentation; https://lnkd.in/dWAnevvQ

❓ Can automated grid sampling improve efficiency in data collection? 🛰️ In their NEW blog, IDinsight’s Data Science and Engineering team members talk about grid-based sampling – a method that enables researchers to sample targets for surveys using geospatial data – and describe the tool they built to make grid-based sampling easier, and more efficient for our teams at IDinsight. 🔗 Delve into the blog here: https://bit.ly/3OYBDSG

Our Paris Data Day for Investors is returning on 30 October! Join us alongside 100+ members of the alternative data community, where we’ll be diving deep into the hottest topics in the alternative data space, including: 🌐 What will 2025 look like for alternative data? 📉 Tackling alpha decay and the lasting value of alternative data 🤖 When machine learning and alternative data collide 💡 Vendor use cases and new product pitches To secure your place, email: 👉 events@neudata.co #neudatadataday #alternativedata #paris

After a year of fascinating research into AI, I'm excited to share some insightful findings. I am pleased to share an article detailing my research into Large Language Models (LLMs) over the past year. This piece presents my primary research and development efforts, focusing on the application of LLMs to enhance human efficiency and automate significant portions of the labelling pipeline. The work explores how we can maintain high labelling quality while improving overall productivity, potentially offering valuable insights for the field of AI-assisted data labelling. I invite you to read the article for a deeper understanding of this intriguing area of study. https://lnkd.in/gJ7XcSBq As I share this work, I would also like to announce that I am leaving Cyabra on good terms. I want to express my gratitude to this remarkable company for an exceptional period of growth and learning. My sincere thanks go to my direct manager, Liza Miller,PhD , for her mentorship and unwavering support, and to Ido Shraga for providing the innovative working environment that made this research possible. I'm also deeply grateful to the entire Cyabra family for their collaboration and friendship. Cyabra is an outstanding company with an important mission, and I am proud to have been a part of it. As I move on to new opportunities in the AI and machine learning field, I will carry with me the valuable experiences and insights gained during my time here. I'm excited about the potential to apply this knowledge in new and challenging contexts. I'd love to hear your thoughts on this research. Feel free to comment or reach out to discuss further! #LargeLanguageModels #AIResearch #MachineLearning #DataScience #AI #NLP

🛰️ To build a comprehensive picture of industrial activity at sea, our machine learning engineers match AIS data against other data types — including satellite imagery. This helps determine if what they are seeing is consistent with what the tracking data reveals. 👤 It’s a complex process. Our senior machine learning engineer, Timothy Hochberg, explains how he approached this work for the new study published in Nature. More features to follow! Dive in below. More features to follow! 🌐 English: bit.ly/41UlPWk 🌐 Spanish: bit.ly/3TRaSms

I'll be speaking at GOTO Chicago on performance techniques in Jepsen. It's gonna be a weird collage of parallelism, pure functions, immutable data structures, and deforestation with low-level techniques like bitsets, avoiding sharing between threads, packing structures into mutable arrays, dynamic compilation of primitive boxes, and macro iteration magic. October 21 & 22nd. https://lnkd.in/d3yedtc6

Almaden Genomics’ g.nome® is the omics data science solution that provides scientists with an end-to-end approach to life sciences discovery. With g.nome, you can simply: 🔎 Select the right datasets for your research using simple queries. ✏️ Pick one of our pre-built, fully modifiable workflows to get quick results and visualizations. 📊 Dig deeper with associated tertiary analysis for a fuller understanding of underlying factors. g.nome enables you to modify any steps along the way with complete traceability and reproducibility of your results. Learn more: https://lnkd.in/dhxj7Z5r

I am excited to be presenting “Spatial Transfer Learning for Estimating PM 2.5 for Data-poor Regions” at #ecmlpkdd2024 today. -- We found that transferring between two distant regions is affected by spatial and semantic dependencies present in the data. -- We introduce a new feature in the data called Latent Dependency Factor (LDF) to capture these dependencies.  -- Our experiments show an improvement of over 19.34% over the baselines. The camera-ready version of the paper can be found at: https://lnkd.in/e4dqQ7gr Thanks to my co-authors: Yongbee Park, Jianzhao Bi, Suyash Gupta, Andreas Züfle, Avani Wildani, and Yang Liu

How important are experiments in verifying your models? The short answer is... very! 🎯 However, if you want a more detailed, deep dive into GeoLift and experiments, then listen back to this afternoon's episode of Marketing Measurement Matters. We were joined by Milosz Bolibrzuch, Data Science Lead at Twigeo and we discussed everything from the iOS 14 impact on Twigeo's approach to marketing measurement, some Bayesian vs. frequentist chat, the importance of not solely relying on MMM and the benefits of verifying your results with experiments. 🚀 Catch up now: https://lnkd.in/eMBEngDS