How do self-driving cars see the environment? Semantic segmentation is to display the same type of objects in the picture with a label(color). My article in Towards Data Science explains U-Net architecture and includes a real-world example in #Python towardsdatascience.com/semantic-segme…

Dioxin, which is a carcinogenic substance in breast milk, can be present in breastfed babies even after many years. The amount of dioxin should be minimal as the benefits of breastfeeding can't be waived. The variation of dioxin levels in breast milk over the years is shown.

Clustering, an unsupervised learning method, is used in various areas such as •recommender system •image segmentation •DNA patterns •anomaly detection I published in Towards Data Science a hands-on tutorial about clustering types and usage areas in #Python towardsdatascience.com/clustering-typ…

The G-8 countries are responsible for two-thirds of the world's, gross domestic product, that is, they represent approximately 65% of the world economy. The graph illustrates the -Annual working hours vs GDP per capita- for G-8 countries. #dataviz

Boosting, one of the ensemble learning methods, builds multiple learning algorithms sequentially for the best result My article in Towards Data Science explains the boosting methods, one of the favorite methods of Kaggle competitions, with #Python implementation towardsdatascience.com/an-overview-of…

XGBoost Is All You Need Deep Neural Networks and Tabular Data: A Survey arxiv.org/pdf/2110.01889…

- How does PCA reduce dimensionality? - How to calculate PCA with the statistical approach from scratch? My blog post published in Towards Data Science explains the theoretical part of PCA including PCA types by supporting it with practical studies in #Python towardsdatascience.com/comprehensive-…

How to Select Loss Function and Activation Function for Classification Problems by Ibrahim Kovan buff.ly/3vnTrw5

How to Make Artistic Images with Neural Style Transfer by Ibrahim Kovan buff.ly/3EB0bus

Why/When you should consider geodesic distance for data visualization and dimensionality reduction? My article in Towards Data Science explains Isometric Mapping(ISOMAP), and includes interpretation of the visualized dataset that reduced dimensionality by ISOMAP towardsdatascience.com/preserving-geo…

A machine learning algorithm that you should study: PCA → Principal Component Analysis. An unsupervised learning algorithm. This picture is Robust PCA, a variation of PCA. Look at how we can use it to remove the background. ↓

And the deep learning vs conventional machine learning for tabular data continues! A new paper looks at 45 mid-sized datasets (10k examples) and finds that tree-based models (XGBoost & random forests) still outperform deep neural networks on tabular datasets. [1/6]

What does Turkish money have to do with quantum computing? The Arf invariant (depicted on the 10 lira banknote) plays a role in the tensor power representation of the qubit Clifford group. A great new paper by Felipe Montealegre-Mora and David Gross: arxiv.org/abs/2208.01688

Machine Learning students try more complex ML models when they wanna improve their results. So they miss the elephant in the room 🐘 ↓

I am pleased to announce that the new book with CRC(Taylor & Francis) and RMIT University researchers has been published Our contribution as a chapter is titled "Forecasting the Energy Consumption Impact of Electric Vehicles by Means of Machine Learning Approaches" amzn.eu/d/9eViNZy