Neural Search - Let's talk about quality
06-13, 14:50–15:30 (Europe/Berlin), Frannz Salon

Context

In the past year the interest in Neural Search and vector search engines increased heavily. They promise to solve multi modal, cross modal and semantic search problems with ease. However, when quickly trying Neural search with off-the-shelf pre-trained models the results are quite disillusioning. They lacking knowledge about the data at hand. In order to explicitly solve model finetuning for search problems we implemented an open-source finetuner. It is directly usable with several vector databases due to the underlying data structure.

Presentation

In our talk we present our methodology and performance on an example dataset. Afterwards, we show how well the approach transfers to other datasets, such as deepfashion, geolocation geoguessr and more. It will give hands-on guidance on how you can finetune a model in order to make your data better searchable.

I enjoy bringing machine learning into production at Jina.ai as an engineering director. The combination of high quality engineering, digging into data and the real-world problem at hand thrills me.

When working in large organizations like SoundCloud, Deloitte and Axel-Springer I learned that the hardest challenges for tech companies are not of technical nature. As a Solution Lead at Jina, I analyze the challenges of our clients and come up with customized solutions. Based on these learnings, I propose changes to our Framework in order to push the quality and accessibility of neural search.