The December 2021 Kirontech bulletin was well received on LinkedIn. Future ones will go out directly on twitter too! linkedin.com/pulse/fraud-it…

The first Kirontech Bulletin of 2022 is out - the topic is Crime -v- Misconduct in insurance claims. How are they different and does it matter from the insurance payer's perspective: linkedin.com/posts/kirontec…

The latest Kirontech Bulletin is out - the topic is handling claims after FWA has been identified. Pre-identification, it's all technical and analysis. Post-identification, it's human factors and soft skills. Hope you find this interesting! linkedin.com/pulse/human-el…

Blood tests are prime source of fraud and abuse: cheap to perform, high mark-ups and little harm to the patient once sample is taken. Unscrupulous clinicians can use these tests to effectively print money. Hope you enjoy this one! linkedin.com/pulse/blood-mo…

My answer to What is a good way to represent time series data as input in deep learning? quora.com/What-is-a-good…

My answer to Is there some work on the theory of activation functions in neural networks? quora.com/Is-there-some-…

My answer to Is it possible to machine learn the different machine learning methods and algorithms? quora.com/Is-it-possible…

My answer to Can I use complex numbers for the states of neural units in deep neural networks? quora.com/Can-I-use-comp…

My answer to What was Leo Breiman trying to convey in his research paper Statistical Modeling - The Two Cultures? W… quora.com/What-was-Leo-B…

My answer to How can we measure the general intelligence of deepmind AI? quora.com/How-can-we-mea…

The coming of age of behavioral machine learning “Graph-based machine learning: Part I” by Sebastien Dery blog.insightdatascience.com/graph-based-ma…

My answer to Is it important to know C++, Algorithms Algorithms and datastructures to have a decent career in Machi… quora.com/Is-it-importan…