CS Colloquium

Irfan Mirza
Director of Enterprise Resilience at Microsoft

Stevenson 1300
Monday, April 21, 2025

As artificial intelligence (AI) continues to evolve and permeate various aspects of industry and life, ensuring public confidence in the underlying technologies that result in AI is paramount. This discussion explores the critical role of responsible AI practices and the ethical citizenship that AI providers must adopt to foster and reinforce trust among users. With the anticipated growth of AI applications across industries, computer scientists and engineers will face an increasing burden of responsibility to create systems that prioritize fairness, accountability, and transparency.

We will discuss the current landscape of public perception regarding AI and the essential principles that are foundational to responsible AI development. As AI becomes more pervasive, the industry must engage diverse stakeholders and promote transparency to address concerns about bias and inequality. Furthermore, we must examine the benefits and challenges of regulatory frameworks in guiding ethical practices and enhancing public trust.

An outcome of this discussion is to highlight actionable strategies for AI providers to demonstrate their commitment to responsible citizenship. Ultimately, this discussion will outline a vision for the future, emphasizing that the path to widespread AI adoption hinges on the industry’s ability to uphold its ethical obligations and build lasting trust with the public.

Dept Event

Stevenson 1300
Monday, April 28, 2025

Short presentations of research carried out by Sonoma State Computer Science Students, and CS awards.

Dept Event

Stevenson 1300
Monday, May 5, 2025

Short presentations of capstone projects carried out by Sonoma State Computer Science Students

Kyle Kabasares
NASA Ames Research Center

Stevenson 1300
Monday, September 8, 2025

Accurately identifying clouds in satellite images is important for studying the Earth’s surface. Traditional methods, such as radiative transfer models and basic machine learning techniques, often struggle to process complex satellite data from the GOES-R system. In this work, I developed a deep learning-based approach to improve cloud masking by combining spatial and temporal information. I built a U-Net convolutional neural network (CNN) to analyze images pixel by pixel, classifying each pixel as CLEAR, PARTLY CLOUDY, or CLOUDY. The model was trained using a set of carefully selected satellite images and achieved nearly 88% accuracy on test and validation data. However, some challenges remain, such as distinguishing clouds from shadows and snow-covered mountains, which can lead to misclassification. Future improvements will focus on optimizing input data, refining the labeling process, and designing better methods to measure model reliability.