Chef’s KISS: Statistics PhD candidate creates new recipe for complex spatial data analysis

Teresa Donnellan — Thu, 06 Mar 2025 20:40:50 +0000

Chef’s KISS: Statistics PhD candidate creates new recipe for complex spatial data analysis Teresa Donnellan Thu, 03/06/2025 - 15:40

In This Story

Ben Seiyon Lee

Jin Hyung Lee, a PhD candidate in the Department of Statistics, in collaboration with his advisor, Assistant Professor Ben Seiyon Lee, has developed novel statistical methodology for analyzing large-scale complex data that can be up to 3,600 times faster than the current standard. He will receive the Korean International Statistical Society (KISS) 2024 Outstanding Student Paper Award for his work and present about it at the American Statistical Association’s Joint Statistical Meetings (JSM) 2025 in Nashville, Tennessee.

The primary application of Lee’s methodology is in the analysis of high-resolution satellite image data, which is inherently complex and high-dimensional. A major challenge lies in predicting values at unobserved locations within the satellite imagery. While traditional spatial modeling methods are computationally intensive and time-consuming, Lee's novel approach offers a solution that is significantly faster. This development has significant implications for various fields, such as atmospheric sciences, ecology, public health, and medical imaging.

Jin Hyung Lee, photo provided

Lee’s research focuses on the application of variational inference to massive spatial datasets using a unique machine learning algorithm. Variational inference is a method used to approximate complex probability distributions when true inference is not possible, making it a powerful method for modeling complex spatial datasets. The proposed approach significantly improves the computational efficiency of spatial data analysis, achieving speeds up to three orders of magnitude faster than existing methods while maintaining comparable accuracy.

Lee's method can be used to predict weather patterns in remote and rapidly changing environments like the Gulf Coast, southern California, and polar regions like Alaska or Antarctica. The ability to process high-dimensional data quickly and accurately is crucial for making timely and informed decisions in these regions, especially when they are at risk of geophysical hazards such as hurricanes, wildfires, or sea ice loss. Additionally, the methodology has potential applications in infectious disease modeling and public health data analysis, where rapid and accurate predictions are essential for effective intervention and policymaking.

Lee's research is part of his broader dissertation work, which focuses on developing and applying variational inference techniques to various statistical models. His goal is to simplify these methods, making them more accessible to non-experts and applicable to a wide range of data types. By doing so, Lee aims to bridge the gap between advanced statistical methodologies and practical applications, enabling more efficient and effective data analysis across different fields.

At the JSM 2025, Lee will have the opportunity to present his research to a large audience of statisticians from around the world. The conference, which attracts over 5,000 participants annually, provides a platform for sharing cutting-edge research, networking with peers, and exploring new developments in the field of statistics. Lee’s presentation will be part of a session hosted by KISS, where he and other award recipients will showcase their work and receive their awards.

Receiving the KISS Outstanding Student Paper Award is a significant milestone in Lee’s academic journey. It not only recognizes his contributions to the field but also provides him with valuable exposure and opportunities for future collaboration. As Lee continues to advance his research, his innovative methodologies and their applications are poised to make a lasting impact on the field of statistics and beyond.

Topics

Department of Statistics

Computational statistics

Image memorability can sharpen our sense of time, finds psychology professor

Colleen Rich — Mon, 22 Jul 2024 15:48:39 +0000

Image memorability can sharpen our sense of time, finds psychology professor Colleen Rich Mon, 07/22/2024 - 11:48

Research by George Mason professor Martin Wiener recently demonstrated that the more memorable an image is, the longer and more accurately its viewers can perceive the passage of time. In a series of experimental studies, participants were more likely to say more memorable images stayed on screen longer, they made that response faster, and they were more consistent about their responses with more memorable images, saying consistently that they stayed on-screen longer.

Martin Wiener. Photo provided

“That study touched on something called memorability, which is the likelihood that you will remember something later,” explained Wiener. “It’s an open question of why there are some things that we just remember very well and some things we forget. Cognitive scientists and computer scientists have been very interested in this question, especially from a machine learning/AI perspective.”

Wiener and colleagues conducted several experiments to determine how the size, memorability, and clutter of images affects our perception of time while viewing them. To examine memorability, they used a database built by computer science researchers at MIT that scores images at varying rates of memorability.

“We were trying to understand our visual sense of time by looking at how different types of images can influence it, and that led to a strong connection to memory that hadn’t really been explored before.”

Wiener and colleagues were also curious if this impacted how likely participants were to remember the images later, so they asked participants 24 hours later if they had seen the images, and they were more likely to remember seeing the images that were more memorable. Additionally, if they had reported seeing the image for a longer period of time (by holding down the space bar), they were even more likely to remember them the next day.

Given the findings of the experimental study with human participants, they used the same test with an artificial intelligence (AI) machine model of the visual system (a recurrent convolutional neural network). This model also reported that it saw the more memorable images for a longer period of time and made this response faster and more consistently. This meant that the findings were not just limited to their experimental study and suggested that there was a mechanism for how the brain processes these images.

An example of the types of photos used in the study using open-source photos. Graphic provided

Wiener explained the significance of these findings. “For some stimuli, the more memorable the images, the brain thinks this is very important. I need to process it as quickly as I can and gather as much information as I can, and in order to do that, I’m going to dilate time a little bit. The experience becomes longer, it becomes expanded. What that translates to is a better encoding into memory so that you can retain that better and remember it better a few hours later, 24 hours later, etc. That’s where we got this connection between memory and time. �鶹�� sense of time seems to be something controllable by the brain and used by the brain to gather information.”

Wiener’s work is unique in that it is looking at very complicated images of scenes and pictures, seeing how those influence the sense of time. Previous researchers studied very simple visual stimuli—different sized squares, or high and low contrast images.

Adding to his work on memory, Wiener runs the Spatial, Temporal, Action, Representation (STAR) Lab at George Mason, which is uncovering how the senses build a perception of time. Students at all levels—from high school volunteers to doctoral students—work with Wiener on studies exploring very short intervals of time—everything from a few hundred milliseconds to multiple seconds to a maximum of about a minute. Some of their work has explored movement and time, which also has critical applications in fields beyond psychology, such as dance and music.

They found that when people are allowed to freely move, their sense of time is sharpened, meaning they perceive the passage of time more precisely. More recent work is exploring how people learn intervals of time—how feedback and learning lets you acquire, understand, and measure intervals of time. Their findings could change how we teach topics and skills related to time.

All this discussion about time may leave you wondering if we can slow down time. As Wiener explains, you can, but generally it’s not a good thing. They tend to be cases of high arousal—very frightening or intense situations. “Usually situations in which time slows down tend not to be pleasant ones,” he said. “They are often frightening or emotional or upsetting, like when you see something scary or something dangerous is occurring.”

He said time also slows down in very boring and monotonous situations, such as waiting on hold or being stuck in traffic. “In those situations, we become more keenly aware of the passage of time, and the more you think about the passage of time, the slower it becomes.”

Why does Wiener study this topic? “The stuff that excites me about time is how little we still know about the way the brain perceives it and measures it but how fundamental it is to everything,” he said. “It’s one of those things…on the surface it might not seem that interesting, but the more you think about it, the more you go wait, how does that happen?”

Topics

Research

College of Humanities and Social Sciences

Department of Psychology

Spatial Analysis

Spatial Analysis

Chef’s KISS: Statistics PhD candidate creates new recipe for complex spatial data analysis

In This Story

Topics

Image memorability can sharpen our sense of time, finds psychology professor

More from the Psychology Department

Read more about Research at Mason

Topics