Pymc-learn is a library for practical probabilistic machine learning in Python. It provides a variety of state-of-the art probabilistic models for supervised and unsupervised machine learning. It is inspired by scikit-learn and focuses on bringing probabilistic machine learning to non-specialists. It uses a syntax that mimics scikit-learn. Emphasis is put on ease of use, productivity, flexibility, & performance.
Last month, I gave a presentation titled Introduction to Probabilistic Machine Learning using PyMC3 at two local meetup groups (Bayesian Data Science D.C. and Data Science & Cybersecurity) in McLean, Virginia. The following is a summary of the concepts we discussed regarding Principled AI.
This study proposed a Bayesian nonparametric framework to capture implicitly hidden structure in time-series having limited data. The proposed framework, a Gaussian process with a spectral mixture kernel, was applied to time-series process for insider-threat detection. The proposed framework addresses two current challenges when analyzing quite noisy time-series having limited data whereby the time series are visualized for noticeable structure such as periodicity, growing or decreasing trends and hard coding them into pre-specified functional forms.
Incase you missed it, here is a recording of my talk on Introduction to Probabilisitic Machine Learning at the Las Vegas R & Data Science Meetup groups. I introduced probabilistic machine learning and probabilistic programming with Stan. I discussed the basics of machine learning from a probabilistic/Bayesian perspective and contrasted it with traditional/algorithmic machine learning
Bayesian optimization provides a principled approach to automate hyperparameter tuning.