Suggestions
Abhishek Mungoli
Senior Data Scientist at Walmart Global Tech India - IIIT-H CSE
Abhishek Mungoli is a seasoned Data Scientist with a strong background in Artificial Intelligence, Algorithms, Big Data, and Machine Learning, complemented by a comprehensive knowledge of Computer Science across different domains. He excels in solving complex problems related to Machine learning and Optimization, with a focus on Retail-specific challenges like Forecasting, Markdowns, Price-elasticity, Item-embeddings, Item-linkage, and Price Investment.
Abhishek is passionate about implementing scalable Machine Learning models and actively engages in knowledge sharing through various mediums including blogs, talks, and meetups. He is dedicated to simplifying intricate concepts and enjoys problem-solving, data science, product development, and optimizing solutions. In his leisure time, he finds relaxation in exploring new places and engaging in workouts.
Abhishek's educational background includes a Bachelor of Technology (BTech) in Computer Science from North Eastern Regional Institute of Science and Technology (NERIST), a Masters of Technology (MTech) in Computer Science from the International Institute of Information Technology, and prior studies at Kendriya Vidyalaya.
Professionally, Abhishek Mungoli has made significant contributions as a Senior Data Scientist at Walmart Global Tech India. Prior to this role, he held positions as a Data Scientist and Statistical Analyst at the same organization. He has also gained valuable experience during an internship as a Trainee at Defence Research and Development Organisation (DRDO).
Highlights
Recommender systems have become an integral part of our daily lives, from online shopping to social media to entertainment platforms.
I've curated a diverse and extensive video playlist within my channel, tailored specifically for individuals who have a keen interest in the field of Recommender Systems.
📌Netflix's Calibrated Recommendations youtu.be/DOWXNrBpO4w 📌Netflix's Unified Recommendation ML Model: youtu.be/OKmv9sUrvk8 📌Evolution of Recommendation Systems | Instagram Pinterest Twitter youtu.be/lgoyJn7MsH8 📌Multi-Armed Bandit Startegies: youtu.be/2A5f3GrX0dA 📌Overcoming Biases for a Better Recommender System | How Tech-Titans Combat Recommender System Biases youtu.be/oGb_mIdO0tA 📌Building Scalable Query-Item Two-Tower Model based Retrieval System youtu.be/o-pZk5R0TZg 📌IIIT Sricity Special: Navigating Twitter and Instagram Recommender Systems youtu.be/PaDsiJCPCXQ 📌Meituan's Game-Changing Two-Tower Retrieval Recsys Model | KDD'21 youtu.be/UhpbTSbi3lI 📌Innovative Linkedin's Deep-leaning based CTR Modeling: The Deep, Wide, and Shallow Towers Explained youtu.be/7l0HLYVFEuU 📌 Lessons from Instagram’s Cutting-Edge Recommendation System youtu.be/Myna6rnmCG8 📌 How Twitter recommends Tweets: Inside Twitter's Recommendation Algorithm youtu.be/IhGq9jgcxFM 📌 Model Based Account Recommendations @ Twitter | Recommendation System | Retrieval & Ranking | RECSYS youtu.be/Xqo8fwgjxW4 📌 Train your own Product Embeddings | Embeddings across Domains | Training on 1.02 Billion rows on GPU: youtu.be/DN4S96oHRhE 📌 Approximate Nearest Neighbour and a popular Library Annoy for k-Nearest Neighbour Search: youtu.be/DSQOrBTqmYA 📌 Approximate Nearest Neighbor and Product Quantizer for k-Nearest Neighbor | Embeddings Search: youtu.be/50PNumB7s3U 📌 PID Controller to add a layer of diversity to the recommender systems: youtu.be/laTxgnzjfR0 📌 Advertiser Recommendation Systems at Pinterest | Machine Learning & Algorithm-driven Ads Platform: youtu.be/KEiBJ0fal3Q 📌 Leveraging Pre-trained Bert to generate Entity Embeddings across domain: youtu.be/v-0J7o-nDBE 📌 How eBay Built a Next-Generation Language Model to Revolutionise Online Shopping youtu.be/h51nbWr7feo
#datatrek #datascience #machinelearning #statistics #deeplearning #ai
Netflix is a cinematic universe where binge-worthy originals and a galaxy of shows and movies converge into personalized entertainment escape.
In this video, we delve into a Netflix paper on calibrated recommendations. The concept behind calibrated recommendations is to ensure that the recommended list of items reflects various (past) areas of interest of a user with their corresponding proportions.
It is observed that modern recommendation systems, built on point-wise or pair-wise strategies, can be biased towards the major category of interests, overshadowing the user's lesser interests.
To address this, the Netflix team has defined a metric to quantify calibrated recommendations and devised a simple yet efficient reranking greedy strategy to balance relevance and calibrated recommendations. The paper also compares calibrated recommendations with diversity, outlining differences, and tweaks the proposed algorithm to account for diversity. It's a beautifully written and well-summarized paper.
Please find out more in this detailed video on the topic on my channel.
Video Link: youtu.be/DOWXNrBpO4w Channel Link: youtube.com/@datatrek
#datatrek #datascience #machinelearning #statistics #deeplearning #ai
