Posted on

The Value of Machine Unlearning for Businesses, Fairness, and Freedom

Our work as data scientists is often focused on building predictive models. We work with vast quantities of data, and the more we have, the better our models and predictions can potentially become. When we have a high-performing model, we continue to retrain and iterate, introducing new data as required to keep our model fresh and free from degrading. The result is that the model’s performance level is largely retained and we, therefore, continue delivering value for users. 

But what happens if restrictions around a data set or individual data point are introduced? How then do we remove this information without compromising the model overall and without kicking off potentially intense retraining sessions? A potential answer that is gaining interest, and that we would like to explore, is machine unlearning. 

Posted on

A Primer on ML and Jupyter Notebook

Recently, I was working on an edge computing demo[1] that used ML (machine learning) to detect anomalies for a manufacturing use case. While I had a generic understanding of what ML is, I lacked the practitioner's understanding of how to use it. Similarly, I’d heard of Jupyter Notebook and was vaguely aware that it was connected with ML, but didn’t really know what it was and how to use one. This article is geared towards people who just want to understand ML and Jupyter Notebook. There are plenty of great resources available if you want to learn how to build ML models.

Caution: If you’re a data scientist then this article is not for you! We’ll be using very simple analysis techniques to serve as a teaching aid. 

Posted on

Machine Learning Model Analysis Using TensorBoard

Machine Learning is growing by leaps and bounds with new neural network models coming up regularly. These models are trained for a specific dataset and are proven for accuracy and processing speed. Developers need to evaluate ML models and ensure that they meet specific threshold values and functions as expected before deployment. There is a lot of experimenting going into improving the model performance, and visualizing differences become crucial while designing and training a model. TensorBoard helps visualize the model, making the analysis less complicated, as debugging becomes easier when one can see what the problem is.

General Practice to Train ML Models

The general practice is to use pre-trained models and perform Transfer Learning to re-train the model for a similar set of data. In a technique called Transfer Learning, a neural network model is first trained on a problem similar to the one that is being solved. One or more layers from the trained model are then used in a new model trained on the problem of interest.

Posted on

Machine Learning in Real-Time vs Rules-Based Detection

Unforeseen Levels of Increased Online Purchasing Leads to Higher Payment Fraud

The unparalleled surge in online purchasing during COVID-19 has led to a 40% rise in online retail fraud attempts, according to the New York-based Fraud.net. Payment fraud had, anyway, been increasing in recent years, with the rise of e-commerce. For instance, online shopping fraud grew by 30% in 2017 – twice as fast as e-commerce sales. A recent True Cost of Fraud Study by LexisNexis found that online credit card fraud is estimated to reach $32 billion in 2020.

Moreover, the pandemic has pushed online shopping to a whole new level.

Posted on

6 Metrics You Need to Optimize for Performance in Machine Learning

6 Metrics to Optimize Performance in Machine Learning

There are many metrics to measure the performance of your machine learning model depending on the type of machine learning you are looking to conduct. In this article, we take a look at performance measures for classification and regression models and discuss which is better-optimized. Sometimes the metric to look at will vary according to the problem that is initially being solved.

You may also like:  Machine Learning Models: Deployment Activities

Examples of metrics in Machine Learning