Suvrakamal Das
Machine Learning Engineer
Kolkata, India
Actions
Suvrakamal Das currently is an ML Engineer. With experience as a SciPyConf presenter and a HackTheNorth '23 Scholar, Suvrakamal is dedicated to advancing AI and machine learning.
Links
Area of Expertise
Topics
Multilingual AI: Trans-Tokenization and ZenML for Better Fine-Tuning
Ever wondered why fine-tuning machine learning models for multilingual applications often leads to subpar results? The problem lies in token representation limitations that hinder effective adaptation across different languages. The solution? Trans-tokenization.
In this talk, I will explore how trans-tokenization addresses these challenges by aligning token embeddings between languages, significantly improving adaptation and reducing performance gaps in low-resource languages. I will demonstrate how this technique, combined with ZenML, enhances the fine-tuning process, resulting in more accurate, efficient, and scalable models within the CNCF ecosystem.
Using an English-to-Dutch translation task as a case study, I’ll walk you through the integration of trans-tokenization with ZenML, offering practical steps to push the boundaries of multilingual AI in cloud-native applications. This talk is perfect for those looking to optimize models for global deployment.
Unmasking Hidden Messages: Leveraging Machine Learning to Detect Embedded Marketing in Ads
In the ever-evolving world of digital marketing, embedded or subliminal messages in advertisements have become a sophisticated strategy to influence consumer behavior. This session will introduce a novel machine learning model designed to detect such embedded marketing cues in ads, providing a new layer of transparency in the advertising industry.
We will delve into the technical aspects of the model, discussing the algorithms used, the training process, and the performance metrics. We'll also showcase a live demonstration of the model in action, revealing how it identifies and interprets hidden marketing messages.
Attendees will gain insights into the intersection of machine learning and digital marketing, understanding how AI can be used to demystify advertising strategies and promote ethical marketing practices. Whether you're a data scientist, a digital marketer, or simply an AI enthusiast, join us to explore this fascinating application of machine learning.
Mamba Models a possible replacement for Transformers?
Transformers, the backbone of many state-of-the-art AI chatbots, excel at tasks like text completion but face challenges with scalability and factual accuracy in extensive contexts. Mamba, a novel architecture inspired by state-space models, addresses these limitations. Mamba offers linear scaling, high accuracy, and faster computation compared to Transformers.
This session explores the details of Mamba, showcasing Mamba's strengths like scalability and accuracy that can be leveraged to automatically generate summaries or aggregate information from large datasets more effectively than existing methods.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top