Martin Hořeňovský
Researcher at Pex
Actions
Martin Hořeňovský is currently a researcher at Pex. He has taught modern C++ at Czech Technical University in Prague, and maintains Catch2, a popular C++ unit testing framework, in the little free time he has left.
Links
An Introduction To Floating Point Math
Many programmers have to deal with floating-point numbers, but often they have only a very superficial understanding of how they work and what it means for the code they write. This issue is further exacerbated by a lot of advice on the internet.
We will go over the very basics of floating-point numbers (as defined by IEEE 754), such as their representation, their correspondence to the scientific notation, what are their limitations, what errors we can expect when using them to do math and also how to compare two numbers.
We will finish by looking at cases where our newfound understanding of floating-point math breaks down, such as using different C++ platforms, different contexts, and other fun things.
5 Years Of Teaching C++: A Retrospective
Before COVID, I spent 5 years teaching C++ to students at my university. I went in with many opinions on teaching in general and C++ in particular, and I still have many opinions on teaching. Just not always the same ones.
In this talk, I want to share some hard-won experiences from my time teaching a C++ course to undergrads, and various things that I would do differently in hindsight. I hope these might be useful to others who look to teach C++ in primarily classroom setting.
Catching up with Catch2: Changes recent and future
Catch2 is a popular unit testing framework with a new major release is coming soon. In this talk, we will go over what has changed in Catch2 recently, and what further changes you can expect as a part of the next major (v3) release.
First, we will take a look at the runtime and compile-time improvements that the new major version brings. Then we will take a look at other changes, including changing the distribution model to a proper library, and new features, such as support for multiple reporters, improved matcher framework and other miscellaneous changes.
Solve hard problems quickly using SAT solvers
What do robot motion planning, hardware design and verification, software verification, model checking, cryptanalysis and bioinformatics have in common? All of them often rely on SAT solvers. This talk will show you how to solve your own hard problems quickly, by leveraging the decades of research that went into the performance of modern SAT solvers.
We will take a look at what the boolean satisfiability (SAT) problem means, and what kind of problems can be converted into SAT. We will then go over examples to show how real-world problems are expressed as SAT, starting with simple examples and ending with a real world problem we have been working on the last few years.
We will also cover why SAT solvers should be especially interesting to C++ developers, how to effectively drive a SAT solver using C++, how we can transfer domain-specific knowledge to the solver and which factors correlate with being able to solve problems quickly.
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top