DataHour: A Simple Guide to Deep Metric Learning

About the DataHour:

Deep metric learning is a technique that focuses on determining the similarity or dissimilarity between data through the use of a distance metric. One approach to deep metric learning is the use of Siamese networks. When training a Siamese network for deep metric learning, two common loss functions are used: Contrastive Loss and Triplet Loss. 

In this DataHour session, we will learn about training a Siamese network using the Triplets loss function, with the sample image dataset. Which can be further used to build a visual search engine, object reidentifications, and face verification tasks.  

Prerequisites: The zeal for learning new technologies, and good to have a conceptual understanding of how neural network works. 

Who is this DataHour for?

• Students & Freshers who want to build a career in the Data-tech domain.
• Working professionals who want to transition to the Data-tech domain.
• Data science professionals who want to accelerate their career growth

Speaker:

Udaya A S

Data Scientist at Jio

Udaya is a highly skilled data scientist at Jio, with a strong track record of delivering high-quality, data-driven solutions that meet business needs. With five years of experience in the field, Udaya specializes in developing recommendation engines and computer vision models for production-level applications. He has deep expertise in data pre-processing, feature engineering, modelling, and deployment, which has allowed him to successfully deliver complex projects across a range of industries. Udaya's ability to develop innovative solutions, combined with his attention to detail and analytical skills, make him a valuable asset to any team.

Connect with him on Linkedin