Aug 31, 2022Collapsing Solutions in Self-Supervised LearningIn the previous post I explained how self-supervised learning has been established as a decent method for unsupervised representation learning. I dicussed pre-text task learning, contrastive learning, and touched upon a few non-contrastive learning methods. In this blog post, I aim to get deeper into similarity learning, which is a…Deep Learning4 min readDeep Learning4 min read
Aug 28, 2022Self-Supervised LearningSelf-supervised learning (SSL) is rapidly closing the gap with supervised methods. Recently, Facebook AI Research (FAIR), one major player in broadening the horizon of self-supervised learning, introduced SEER. SEER is a 1.3B parameter self-supervised model pre-trained on 1B Instagram images that achieves 84.2% top-1 accuracy on ImageNet, comfortably surpassing all…Machine Learning11 min readMachine Learning11 min read
Oct 22, 2018Collaborative Filtering In Simple LanguageNowadays everybody somehow knows what recommender systems are. But do we know how do these magical systems work and how they understand what products we might like? In last few years, large e-commerce websites as Amazon caused a massive growth in the number of users’ choices. These websites provide more…Machine Learning4 min readMachine Learning4 min read
