Conference Schedule

Light lunch will be provided

This session is open to principal investigators (PIs) only. Over the last few years, the number of large-scale initiatives advancing accelerated discovery has increased substantially. How do we ensure that these initiatives build on each other's success, work collaboratively, and create shared resources? In this session, we will discuss opportunities for (1) deepening collaborative research efforts and (2) developing an open-source hardware, software, and data community.

Virtually every real-world chemistry and materials informatics task involves optimizing multiple properties of interest, weighing trade-offs of between experiment value and cost, and optimizing many tunable parameters at once. While traditional design of experiments is often used, Bayesian optimization is a dramatically more efficient alternative in many cases. In this workshop, participants will learn about why we need Bayesian Optimization and about topics such as multi-objective, multi-fidelity, and high-dimensional Bayesian optimization.

In Part II of real-world Bayesian optimization, participants will form groups and work collaboratively in a hackathon-style breakout room. Each group will follow a Python tutorial of choice from the list of topics in Part I and adapt it to a new materials optimization task. The Python tutorials can be run in Google Colab, so no installation is required.

This session is open to principal investigators (PIs) only. Over the last few years, the number of large-scale initiatives advancing accelerated discovery has increased substantially. How do we ensure that these initiatives build on each other's success, work collaboratively, and create shared resources? In this session, we will discuss opportunities for (1) deepening collaborative research efforts and (2) developing an open-source hardware, software, and data community.

Computer vision is a powerful tool for interpreting our world through digital images. This three-part workshop helps students learn how to recognize and categorize microorganisms using computer vision. Part I covers image classification using convolutional neural networks in Google Colab, a web-based coding platform. Part II introduces object detection using Mask R-CNN. Part III involves object detection exercises using SSD-mobilenet on the Jetson Nano platform. Participants are encouraged to join the AC discord channel prior to attendance and bring a laptop.

Software development can be rewarding yet time-consuming and frustrating for developers and users: files get lost or out-of-sync, mistakes in a single line of code can cause catastrophic errors, bugs go unnoticed for days and weeks, and installing a package takes hours of troubleshooting. Applying best practices comes with an initial startup cost, but it maximizes effectiveness and minimizes frustration. Participants will apply practical tools and guidelines for software development for topics including version control, tests, documentation, debugging, packaging. SDL-specific topics include workflow orchestration and cloud computing. Participants are encouraged to join the AC discord channel prior to attendance and bring a laptop. This workshop is sponsored in part by Amazon Web Services.