Faculty Supervisor: Donna French

Description: Interested in one of the OurCS@DFW Workshops, but don't have any Python experience? Python is an easy language for anyone to learn even for coding beginners. Unlike some programming languages, Python's use is expanding and covers a range of programming needs, from basic to advanced. It is frequently used in data science and machine learning which makes it an excellent choice for those interested in those topics. Python is considered a general-purpose language and offers multiple uses in web, software, game development, and more. This tutorial takes place one week before the OurCS@DFW workshops. If you are interested in learning just enough Python to get you started on your own Python journey or to prepare you for one of the workshops, then this tutorial is for you.

Expected coding experience: Basic-level coding experience in some other language - no experience in Python itself is expected. This tutorial will be most beneficial to those with some type of programming background. This background does not need to be extensive - a high school level Java class for example would be sufficient. Due to the limited amount of time, the tutorial, as an example, will focus on how to implement a for loop in Python as opposed to what is a for loop.

[SUDS] Suds & Sensors: Sensor-driven Latent Space Cartography for Cold-Processed Soapmaking

Faculty Supervisor: Cesar Torres

Description: Discover how traditional soap-making can be paired with modern technology in this interactive workshop. We'll use tools like rheometers (which measure how materials flow) and embedded sensors to track how cold-processed soap changes from a strong, caustic mixture into a fully cured bar. As we collect real-time data from these sensors, we'll create a visual “map” (sometimes called latent space cartography) that helps us see and understand the soap's changing properties. By building what's known as an embedding space—a simplified way of showing lots of different sensor readings at once—you'll be able to watch how soap mixtures evolve over time. We'll also explore how smart tools can guide both beginners and experts by providing instant feedback that may improve safety, efficiency, and creativity. Along the way, we'll discuss how slow, step-by-step transformations can be tracked and displayed, helping us design interactive systems that fit the needs of makers at different skill levels. In the end, you'll gain hands-on experience with a centuries - old craft while exploring cutting-edge, sensor-driven design. This blend of soap-making and technology offers a unique look at how human-computer interaction can shape—and be shaped by—practical, material-based practices.

Accepts high school & community college students (18+ only, due to wavier required for handling caustic chemicals)

Expected coding experience: Having taken an introductory programming course is beneficial but not strictly required.

[PatRec] Pattern Recognition Using Machine Learning

Faculty Supervisor: Vassilis Athitsos

Description: In this workshop, participants will learn some basic techniques for using machine learning to recognize patterns such as letters, faces, or spam messages. We will briefly describe machine learning methods such as neural networks, nearest neighbor classification, and decision trees. Projects will be implemented in Python, using the online Google Colab platform. A brief introduction to the Keras and Tensorflow programming libraries will be given, that will describe how to quickly design and train neural networks for some of these applications.

Expected coding experience: Basic-level coding experience in Python.

[IA-CAV] Design the Intelligent Hub for Connected and Autonomous Vehicle Networks

Faculty Supervisors: Dianqi Han, Xiaojun Shang

Description: This workshop delves into the transformative potential of cloud and edge intelligence in shaping the future of connected and autonomous vehicles (CAVs). As these vehicles navigate complex environments, advanced AI models like Vision Transformers (ViTs) enable them to interpret their surroundings with unprecedented accuracy. To integrate CAVs with such large models, Cloud/Edge computing and networks are essential for addressing the resource and energy limitations of the vehicles' onboard systems. Through hands-on experience in the CARLA simulator, you will start with a basic onboard autonomous driving agent navigating a virtual city and then enhance your design using SOTA techniques, such as ViTs and Vision-Language Models (VLMs), to ensure safety and efficiency in heavy traffic and rare critical scenarios. This workshop offers practical insights into seamlessly integrating AI, computing, and networks within the vehicle-edge-cloud ecosystem to drive the future of intelligent mobility.

Expected coding experience: None

[ImmerSec] Safeguarding Immersive Computing

Faculty Supervisor: Jiayi Meng and Faysal Hossain Shezan

Description: Immersive computing, including Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR), are revolutionizing how individuals interact with each other and technologies. However, they also bring unique security and privacy challenges, considering VR headsets and AR glasses are equipped with multiple RGB/LiDAR sensors, which continuously capture users' behaviors and their surroundings and can expose sensitive user information. This workshop introduces you to the exciting field of immersive computing while uncovering the potential risks these technologies pose to user data and privacy. You will engage in hands-on activities to explore real-world vulnerabilities on today's VR/AR devices, learn how to assess threats, and develop strategies to protect user security and privacy

Expected coding experience: Basic-level coding experience for Python

[GNN] Hands-on Introduction to Graph Neural Networks

Faculty Supervisor: Yuede (YJ) Ji

Description: "Hands-on Introduction to Graph Neural Networks" is an engaging workshop designed to introduce high-school, undergraduate, and prospective graduate students to the fascinating world of Graph Neural Networks (GNNs). This session combines interactive lectures with practical coding exercises to demystify how GNNs work and their applications in solving complex problems on graph-structured data, such as social networks, recommendation systems, and molecular biology. Participants will gain hands-on experience using Python and PyTorch Geometric, learning to build and train GNN models from scratch, regardless of their prior experience with machine learning. The workshop aims to spark curiosity, build foundational skills, and inspire students to explore cutting-edge research and careers in artificial intelligence.

Expected coding experience: Python

[RobustSumm] RobustSumm: Exploring the Power and Pitfalls of Text Summarization Models

Faculty Supervisor: Shirin Nilizadeh

Description: Large Language Models (LLMs) have revolutionized how machines understand and generate human language. One of their most practical applications is text summarization, where lengthy documents are condensed into concise, coherent overviews. Although these models are powerful, they're not without weaknesses. Even small tweaks in what they see or learn, whether by accident or design, can lead to misleading or incomplete summaries. In this workshop, participants will gain hands-on experience exploring both the strengths and limitations of text summarization models. We start by reviewing the fundamentals of how these models process language and generate summaries. Through guided exercises, students will see how test-time manipulations (like slight edits in the input text) can change a model's output in surprising ways. We'll then look at the training process, discovering how a concept known as “model poisoning” (small but intentional changes to the training data) can dramatically influence the final summaries. This hands-on experience will provide students with valuable insights into both the power and limitations of current AI technology in text summarization tasks.

Expected coding experience: A basic familiarity with Python (variables, loops, simple data structures) is recommended.