<< nn-experiments

I might just note some interesting papers here, now that i have a static site renderer. (I'm browsing arxiv.org every other day, for recreational purposes..)

PhishGuard: A Convolutional Neural Network-Based Model for Detecting Phishing URLs with Explainability Analysis

2404.17960

Ha! So when setting up your next phishing urls, make sure this network does not recognize them.

A Comprehensive Survey of Convolutions in Deep Learning: Applications, Challenges, and Future Trends

2402.15490

A thick survey about convolutional nets, focused on the last ten years, with a lot of interesting tables to compare things.

There is also some not-especially-CNN-related stuff of the recent years in there. Basically you can browse through the paper and check if you already know of all these things or not. None of it is explained super-in-depth, though.

TinyLidarNet: 2D LiDAR-based End-to-End Deep Learning Model for F1TENTH Autonomous Racing

2410.07447

Did not know about the F1/10 before. It sure is fun to attend.

This paper describes an RC racer, equipped with a LiDAR depth sensor and a tiny on-board computation device that even handles the training of a 1d convolutional network to control the racer.