At the intersection of two domains: machine learning and data structures lie learned data structures. Learned data structures offer improved performance as compared to their classical counterparts by making use of trends in the input data. We will explore Learned Bloom Filters which build on classical Bloom Filters and offer trade-offs which can be used to achieve optimum performance in some applications.

Bloom Filters

Motivation

A membership query returns a yes/no answer to a question of the form “Is this element \(x\) present in a given set \(S\)?”

Searching algorithms like linear search and binary search can be used to answer this... read more

Neaural Architectural Search for Natural Language

What is NAS?

Deep Learning has made remarkable progress over the years on a number of tasks including Speech and Image recognition , Machine Translation , Automated Driving etc. One very important part of this great progress is due to novel neural architectures. The currently used architectures are mostly developed manually by humans which is often a time and resource consuming process whilst being prone to error. To mitigate this issue, the concept of Neural Architecture Search (NAS) was introduced. NAS automates the design of Artificial Neural Networks and resides within the domain of... read more

As students who didnt have much experience with ML, we started off our journey to implementing SGLD by first familiarising ourselves with the basics of ML. As a first step, we implemented Linear and Logistic Regression models in python from scratch. In doing so, we developed a good understanding of the Stochastic Gradient Descent algorithm, and from here on we proceeded to try to parallelise and distribute it on the server provided. After doing so successfully, we moved on to Bayesian machine learning and began exploring the Naive Bayes classifier as well as Bayesian Linear Regression. We implemented the two... read more