Collaborating Authors


Machine Learning And Easter Egg Hunts


Easter is the quintessential spring holiday, full of vibrant colors, sweets, and family traditions. And yet, it may also be one of the few holidays with a built-in competition: the infamous Easter egg hunt! It usually goes something like this: parents hide colored eggs throughout the yard and kids hunt to try and fill up their baskets before their treasures are scooped up by other seekers. It's the only time of the year when putting all your eggs in one basket is a good thing. As any master egg hunter knows, this is an exercise in pattern recognition and anomaly detection.

Analyzing The Outliers: - AI Summary


In the sphere of data science, anomaly detection is one of the newest buzzwords, but understanding why anomaly detection matters can be challenging. Amazon Lookout for Metrics is a new machine learning (ML) service that processes your business and operational time series data to automatically detect and diagnose anomalies, such as an unusual rise in product sales or an unexpected drop in throughput. The actual Amazon Lookout for Metrics makes it simple for users to diagnose detected data anomalies by grouping related anomalies together and automatically sending an alert warning that helps determine the potential root cause. In the world of data, an anomaly is any point of data that has differences or deviations from the normal data in a dataset. To better understand the different types of anomalies, let's look at examples from the three major types: Point Anomalies, Collective Anomalies, and Contextual Anomalies.

Analyzing the Outliers:


In an increasingly connected world, the connectivity and flow of data and information between sensors and devices creates a tremendous amount of available data. This presents a major challenge for businesses -- how can we process these vast amounts of available data in order to extract valuable information? In the sphere of data science, anomaly detection is one of the newest buzzwords, but understanding why anomaly detection matters can be challenging. Amazon Lookout for Metrics is a new machine learning (ML) service that processes your business and operational time series data to automatically detect and diagnose anomalies, such as an unusual rise in product sales or an unexpected drop in throughput. As official launch partners for Amazon Lookout for Metrics, TensorIoT wants to share how we're using this innovative technology in our solutions.

4 Machine learning techniques for outlier detection in Python


Based on the feedback given by readers after publishing "Two outlier detection techniques you should know in 2021", I have decided to make this post which includes four different machine learning techniques (algorithms) for outlier detection in Python. Here, I will use the I-I (Intuition-Implementation) approach for each technique. That will help you to understand how each algorithm works behind the scenes without going deeper into the algorithm mathematics (the Intuition part) and implement each algorithm with the Scikit-learn machine learning library (the Implementation part). I will also use some graphical techniques to describe each algorithm and its output. At the end of this article, I will write the "Key Takeaways" section which will include some special strategies for using and combining the four techniques.

Machine Learning Engineer - Online Retail Analytics at Apple


Apple's Online Retail Analytics team is looking for a hardworking Machine Learning Engineer who is passionate about crafting, implementing, and operating production machine learning solutions that have direct and measurable impact to Apple and its customers. You will design, build and deploy predictive modeling and statistical analysis techniques on production systems that drive increased sales, improved customer experience for our online customers. Apple has a tremendous amount of data, and we have just scratched the surface in pattern detection, anomaly detection, predictive modeling, and optimization. There are many exciting problems to be discovered and solved and many business owners eager to use data mining. The Apple Analytic Insight team encourages scientists to stay ahead of data science research by attending conferences and working with academic faculty and students.

Anomaly detection using principles of human perception Machine Learning

In the fields of statistics and unsupervised machine learning a fundamental and well-studied problem is anomaly detection. Although anomalies are difficult to define, many algorithms have been proposed. Underlying the approaches is the nebulous understanding that anomalies are rare, unusual or inconsistent with the majority of data. The present work gives a philosophical approach to clearly define anomalies and to develop an algorithm for their efficient detection with minimal user intervention. Inspired by the Gestalt School of Psychology and the Helmholtz principle of human perception, the idea is to assume anomalies are observations that are unexpected to occur with respect to certain groupings made by the majority of the data. Thus, under appropriate random variable modelling anomalies are directly found in a set of data under a uniform and independent random assumption of the distribution of constituent elements of the observations; anomalies correspond to those observations where the expectation of occurrence of the elements in a given view is $<1$. Starting from fundamental principles of human perception an unsupervised anomaly detection algorithm is developed that is simple, real-time and parameter-free. Experiments suggest it as the prime choice for univariate data and it shows promising performance on the detection of global anomalies in multivariate data.



For Octave/MatLab version of this repository please check machine-learning-octave project. This repository contains examples of popular machine learning algorithms implemented in Python with mathematics behind them being explained. Each algorithm has interactive Jupyter Notebook demo that allows you to play with training data, algorithms configurations and immediately see the results, charts and predictions right in your browser. In most cases the explanations are based on this great machine learning course by Andrew Ng. The purpose of this repository is not to implement machine learning algorithms by using 3rd party library one-liners but rather to practice implementing these algorithms from scratch and get better understanding of the mathematics behind each algorithm.

SSD: A Unified Framework for Self-Supervised Outlier Detection Artificial Intelligence

We ask the following question: what training information is required to design an effective outlier/out-of-distribution (OOD) detector, i.e., detecting samples that lie far away from the training distribution? Since unlabeled data is easily accessible for many applications, the most compelling approach is to develop detectors based on only unlabeled in-distribution data. However, we observe that most existing detectors based on unlabeled data perform poorly, often equivalent to a random prediction. In contrast, existing state-of-the-art OOD detectors achieve impressive performance but require access to fine-grained data labels for supervised training. We propose SSD, an outlier detector based on only unlabeled in-distribution data. We use self-supervised representation learning followed by a Mahalanobis distance based detection in the feature space. We demonstrate that SSD outperforms most existing detectors based on unlabeled data by a large margin. Additionally, SSD even achieves performance on par, and sometimes even better, with supervised training based detectors. Finally, we expand our detection framework with two key extensions. First, we formulate few-shot OOD detection, in which the detector has access to only one to five samples from each class of the targeted OOD dataset. Second, we extend our framework to incorporate training data labels, if available. We find that our novel detection framework based on SSD displays enhanced performance with these extensions, and achieves state-of-the-art performance. Our code is publicly available at

Homophily Outlier Detection in Non-IID Categorical Data Artificial Intelligence

Most of existing outlier detection methods assume that the outlier factors (i.e., outlierness scoring measures) of data entities (e.g., feature values and data objects) are Independent and Identically Distributed (IID). This assumption does not hold in real-world applications where the outlierness of different entities is dependent on each other and/or taken from different probability distributions (non-IID). This may lead to the failure of detecting important outliers that are too subtle to be identified without considering the non-IID nature. The issue is even intensified in more challenging contexts, e.g., high-dimensional data with many noisy features. This work introduces a novel outlier detection framework and its two instances to identify outliers in categorical data by capturing non-IID outlier factors. Our approach first defines and incorporates distribution-sensitive outlier factors and their interdependence into a value-value graph-based representation. It then models an outlierness propagation process in the value graph to learn the outlierness of feature values. The learned value outlierness allows for either direct outlier detection or outlying feature selection. The graph representation and mining approach is employed here to well capture the rich non-IID characteristics. Our empirical results on 15 real-world data sets with different levels of data complexities show that (i) the proposed outlier detection methods significantly outperform five state-of-the-art methods at the 95%/99% confidence level, achieving 10%-28% AUC improvement on the 10 most complex data sets; and (ii) the proposed feature selection methods significantly outperform three competing methods in enabling subsequent outlier detection of two different existing detectors.