Published in Towards Data Science·Apr 19, 2019Small questions about “bigger” issues in Machine learningControl theory, linear regression, and models assumptions! — ML Models and data distribution Question 1: Machine learning models have ‘data’ distribution assumptions, right? Answer 1: Not all models! “Statistical modeling: Two Cultures”, highlights an important distinction between two very different approaches to statistical modeling: Stochastic techniques, where a distribution is estimated from successive draws; therefore models assume some underlying distribution.Machine Learning3 min readMachine Learning3 min read
Published in Towards Data Science·Feb 22, 2019How does FastText classifier work under the hood?When linear models outperform complex deep learning models FastText, by Facebook Research, is a library for efficient learning of word representations and text classification. FastText supports supervised (classifications) and unsupervised (embedding) representations of words and sentences. However, the documentation of the FastText package doesn’t provide details about the implemented classifier…Machine Learning2 min readMachine Learning2 min read
Published in Towards Data Science·Feb 9, 2019The power of Brain-Computer interface: use your brain to play your video gameHere we show one application of machine learning and signal processing in Neuroscience; translating thoughts into actions with our Game-Based Brain-Computer interface (BCI). BCI enables direct control…Brain Computer Interface1 min readBrain Computer Interface1 min read
Published in Towards Data Science·Feb 8, 2019The Multiple faces of ‘Feature importance’ in XGBoostBe careful when interpreting your features importance in XGBoost, since the ‘feature importance’ results might be misleading! This post gives a quick example on why it is very important to understand your data and do not use your feature importance results blindly, because the default ‘feature importance’ produced by XGBoost…Machine Learning4 min readMachine Learning4 min read
