Marton Trencseni - Thu 06 January 2022 • Tagged with entropy, physics, spin, glass
I use Monte Carlo simulations to explore the dynamic behaviour of probabilistic spin glasses, specifically how saturation scales with $p$ and $N$. 
Marton Trencseni - Fri 31 December 2021 • Tagged with entropy, physics, spin, glass
This is a continuation of the previous articles on probabilistic spin glasses. I run simulations to understand the scaling behaviour for large spin glasses. 
Marton Trencseni - Sat 25 December 2021 • Tagged with entropy, physics, spin, glass
I run simulations to understand the dynamic probabilistic evolution of these toy models. 
Marton Trencseni - Sat 18 December 2021 • Tagged with entropy, physics, spin, glass
This is a continuation of the previous article on probabilistic spin glasses, with improvements to the simulation code and improved entropy computation. 
Marton Trencseni - Sat 11 December 2021 • Tagged with entropy, physics, spin, glass
I run Monte Carlo simulations on probabilistic spin glasses, a simple mathematical model of magnetized matter with short range interactions. I use entropy to characterize the model's order-disorder transition. 
Marton Trencseni - Mon 29 November 2021 • Tagged with entropy, physics
I derive the Sackur-Tetrode equation for entropy of a monatomic ideal gas. 
Marton Trencseni - Fri 19 November 2021 • Tagged with entropy, physics
I show the first steps of how to arrive at a definition of entropy for a monatomic ideal gas modeled as hard billiard balls. 
Marton Trencseni - Fri 29 October 2021 • Tagged with startups, cocoon, facebook
The idea behind WeToddle came from the Baby Fanclub group we have on Messenger, which has most of our family in it. It turns out some ex-Facebook people had a similar idea in 2019, raised $3M, spent a 2 years on it, and then gave up because it didn’t go anywhere (presumably).
Marton Trencseni - Sun 24 October 2021 • Tagged with entropy
I discuss 4 uses of entropy in Data Science: (i) cross entropy as a loss function for training neural network classifiers (ii) entropy as a splitting criterion for building decision trees (iii) entropy for evaluating clustering algorithms (iv) entropy for understanding relationships in tabular data. 
Marton Trencseni - Mon 18 October 2021 • Tagged with meta
A review and introspect on the first 100 articles written on Bytepawn. 
Marton Trencseni - Sat 09 October 2021 • Tagged with entropy, cross-entropy, joint-entropy, conditional-entropy, relative-entropy, kullback–leibler-diverence
What's the difference between cross entropy, joint entropy, conditional entropy and relative entropy? 
Marton Trencseni - Sat 25 September 2021 • Tagged with entropy, interviews, cross-entropy, physics
What's the entropy of a fair coin toss? What if the coin almost always returns Heads? My recruiter reports that very few candidates can answer these entropy related DS screening questions. 
Marton Trencseni - Sun 19 September 2021 • Tagged with ab-testing, variance, stratification, cuped
I use toy Monte Carlo simulations to demonstrate 5 ways to reduce variance in A/B testing: increase sample size, move towards a more even split, reduce variance in the metric definition, stratification and CUPED.
Marton Trencseni - Sun 05 September 2021 • Tagged with ab-testing, cuped
In this final blog post about CUPED, I will address some questions about CUPED, such as, is correlation between "before" and "after" the same as seasonality?
Marton Trencseni - Sun 15 August 2021 • Tagged with ab-testing, cuped
I use Monte Carlo simulations of A/A tests to demonstrate how Data Scientists can incorrectly skew lift and p-values if they pick-and-choose between reporting traditional and CUPED results after the experiment has concluded.
Marton Trencseni - Sat 07 August 2021 • Tagged with ab-testing, cuped
I use Monte Carlo simulations of conversion A/B tests to demonstrate how CUPED reduces measurement variance in conversion experiments.
Marton Trencseni - Sat 31 July 2021 • Tagged with ab-testing, cuped
I use Monte Carlo simulations of A/B tests to demonstrate CUPED, a method to use historic "before" data to reduce the variance in the measurement of the treatment lift.
Marton Trencseni - Sun 25 July 2021 • Tagged with redacted
I show how looking at historic "before" values in A/B testing can lead to an apparent paradox.
Marton Trencseni - Fri 23 July 2021 • Tagged with data, fallacies
What are the core skills a data scientist needs to sustainably achieve bottom-line impact, without blocking on external help from other roles?
Marton Trencseni - Tue 20 July 2021 • Tagged with modeling, timeseries, prophet, neuralprophet
I compare Prophet and NeuralProphet performance using a toy forecasting benchmark.