У нас вы можете посмотреть бесплатно When music just doesn't match our vibe, can AI help? - Filip Makraduli | Vector Space Talks или скачать в максимальном доступном качестве, видео которое было загружено на ютуб. Для загрузки выберите вариант из формы ниже:
Если кнопки скачивания не
загрузились
НАЖМИТЕ ЗДЕСЬ или обновите страницу
Если возникают проблемы со скачиванием видео, пожалуйста напишите в поддержку по адресу внизу
страницы.
Спасибо за использование сервиса ClipSaver.ru
When the Music we Listen to Just Doesn't Match our Vibe, Can AI Help? by Filip Makraduli Recommendation system using LLMs Abstract: An AI song recommendation tool that tries to understand the vibe and feel of a song. A dataset of songs and their descriptions of moods and vibes is generated by promoting an LLM. Each song's vibe is encoded using a transformer model and saved as a vector index. Then this vector index is queried by similarity using the vector embedding of the user's input and the top 10 matches are returned. Filip's Bio: I began my academic journey as a student of electrical engineering in Skopje, Macedonia. This education, grounded in mathematics and science, provided me with the foundational knowledge to explore various technical fields in my career. After completing my undergraduate studies, I moved to London to pursue a Master's in Biomedical Data Science at Imperial College London. Following this, I joined the Digital and Technology team at Marks and Spencer (M&S). The field of retail, from a data science perspective, is fascinating due to the abundance of valuable data and practical applications. Sometimes this goes a bit under the radar. At M&S, several tech teams work on various machine learning (ML) and artificial intelligence (AI) projects. I had the opportunity to work with amazing people and contribute to different teams, working on tasks such as causal ML and XGBoost models, natural language processing (NLP), and generative AI. Currently working on recommendation systems for outfits. In addition to my professional interests, I have a keen interest in tech startups, entrepreneurial activities, innovative initiatives and ML research, and practical technology applications. It was through my exploration of these areas that I discovered Qdrant, and liked it as a startup and project. I'm pleased to have been invited by Andre to talk at this event! -------------------------------------------------------------------------------- Connect with Filip: / filip-makraduli-55091310a Link to Filip's Project Repo: https://github.com/fm1320/song-vibe Follow Qdrant: / qdrant / qdrant_engine Join the Qdrant Discord server: / discord