events

Ask Maurício for the Zoom link.

Abstract: The success of Deep Neural Networks with image classification prompted researchers to explore the applications of Deep Learning in Medical Imaging and Medical Image Analysis (MIA). Deep Neural Networks have sufficiently demonstrated their capabilities of performing MIA tasks tirelessly and with fewer errors as opposed to their human counterpart. However the challenge of training neural networks using sensitive medical data, without violating the privacy of patients remains an active field of research.

Abstract: In this talk I will go over how we can use Emacs (the mighty text editor) and org-mode (markdown’s big brother) to craft a research workflow. We will look at how we can leverage the power of Emacs and org-mode to capture, store, search and retrieve research data, all in plain text! The talk will also touch upon how org-mode can be used as an environment for literate programming and reproducible research.

A demonstration showing the reliability of serving users recommendations with tradeoff for large music collections, by leveraging diverse Recommender Systems and Evolutionary Algorithms. To sign up for the event you can either fill in this form (https://forms.gle/icuLgPD8gn6Ej19i9) or email s.p.bobde@student.tudelft.nl for the link.

Abstract: Automatically detecting the positions of key-points (e.g., facial key-points or finger key-points) in an image is an essential problem in many applications, such as driver’s gaze detection and drowsiness detection in automated driving systems. With the recent advances of Deep Neural Networks (DNNs), Key-Points detection DNNs (KP-DNNs) have been increasingly employed for that purpose. Nevertheless, KP-DNN testing and validation have remained a challenging problem because KP-DNNs predict many independent key-points at the same time—where each individual key-point may be critical in the targeted application—and images can vary a great deal according to many factors.

While SQL engines are now capable of detecting a large number of syntactic mistakes, most often semantic errors are not detected, which can lead to serious performance issues or even security vulnerabilities being introduced in the system. This thesis proposes a set of 25 validated heuristics together with a new rule-based static analysis tool for detecting the most common types of semantic bugs in SQL queries, based on evidence from previous research.

Abstract: In this talk, I will first explain some of my previous studies as a Ph.D. student. In these studies, we introduced multiple techniques to improve automated test generation using various available information sources, such as the source code and manually-written tests. I will also talk about some of the lessons that I learned in my journey as a Ph.D. student. Finally, I briefly talk about the new topics that I am currently working on as a PostDoc.

Dear all, Join us on Zoom for the next SERG Reading Club 📖📚! In this session, we read and review the paper “FlakeFlagger: Predicting Flakiness Without Rerunning Tests”, which is presented at the ICSE'21. Wanna join & you are not part of our Mattermost? Send an email to c.e.brandt /at/ tudelft.nl to get the Zoom link 😊

Abstract: In this talk, we will introduce SynTest - A framework for generating synthetic tests for programs based on JavaScript. One of the main use cases we have been working on lately, is smart contract testing for Solidity. It is very important for smart contracts to be thouroughly tested since after these contracts are deployed, it is impossible to change or update them. Hence any bugs or security exploits in the code will be in the contract for the duration of its lifetime.

The most common reason for Continuous Integration (CI) build failures is failing tests. When a build fails, a developer often has to scroll through hundreds to thousands of log lines to find which test is failing and why. Finding the issue is a tedious process that relies on a developer’s experience and increases the cost of software testing. Providing CI build test results with additional context in the developer’s local development environment could help solve failing tests more quickly.