Matthew Korban

Matthew Korban

Auto-generated profile, based on publicly available information

Matthew Korban

Department of Electrical and Computer Engineering, University of Virginia, Charlottesville, VA, USA

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 March 2024 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
MKMatthew Korban
Matthew Korban
PYPeter Youngs
Peter Youngs
SAScott T. Acton
Scott T. Acton
This paper presents a novel spatiotemporal transformer network that introduces several original components to detect actions in untrimmed videos. First, the multi-feature selective semantic attention model calculates the correlations between spatial and motion features to model spatiotemporal interactions between different action semantics properly. Second, the motion-aware network encodes the locations of action semantics in video frames utilizing the motion-aware 2D positional encoding algorithm. Such a motion-aware mechanism memorizes the dynamic spatiotemporal variations in action frames that current methods cannot exploit. Third, the sequence-based temporal attention model captures the heterogeneous temporal dependencies in action frames. In contrast to standard temporal attention used in natural language processing, primarily aimed at finding similarities between linguistic words, the proposed sequence-based temporal attention is designed to determine both the differences and similarities between video frames that jointly define the meaning of actions. The proposed approach outperforms the state-of-the-art solutions on four spatiotemporal action datasets: AVA 2.2, AVA 2.1, UCF101-24, and EPIC-Kitchens.

2 November 2025 in Pattern Recognition
Verified
Journal Article
Closed access
MKMatthew Korban
Matthew Korban
PYPeter Youngs
Peter Youngs
SAScott T. Acton
Scott T. Acton


24 October 2025 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
MKMatthew Korban
Matthew Korban
PYPeter Youngs
Peter Youngs
SAScott T. Acton
Scott T. Acton
We present a causal state space model (CSSM) for video understanding that couples a learned causal DAG with latent state dynamics. Latent factors form DAG nodes, enabling explicit cause–effect modeling over time; the state-space form provides efficient sequence inference, while the graph adds interpretability and robustness to distribution shifts. We learn the latent graph and inject its adjacency into the transition operator. On HMDB-51, UCF-101, and HAR, CSSM improves accuracy over strong baselines and supports counterfactual reasoning about video events.

20 March 2025
Verified
Book Chapter
Closed access
PDF file
PYPeter Youngs
Peter Youngs
JFJonathan K. Foster
Jonathan K. Foster
PCPatricia “Scout” Crimmins
Patricia “Scout” Crimmins
GWGinger S Watson
Ginger S Watson
MKMatthew Korban
Matthew Korban
SAScott T. Acton
Scott T. Acton
In this chapter, we first describe the nature and consequences of different activity structures in elementary classrooms and the need to consider the nature of instructional activities within and across activity structures. Second, we discuss the main components of the Artificial Intelligence for Advancing Instruction (AIAI) teacher dashboard and explain how they align with our neural network analyses of activity structures and instructional activities in videos of elementary mathematics and reading/language arts instruction. The dashboard displays videos of lessons; provides information about the frequency and duration of activity structures within and across lessons; and prompts teachers to reflect when interacting with and interpreting the automated classroom data visuals. Third, we report results from a pilot study that included six mid-career elementary teachers in Virginia. The pilot study explored teachers’ perceptions of and experiences with the dashboard, including teachers’ responses to detailed information about their use of activity structures during observations of mathematics and reading/language arts lessons. Finally, we address the next steps in developing a dashboard that will help teachers notice features of equitable, rigorous instruction.

5 November 2026 in npj Artificial Intelligence
Verified
Journal Article
Open access
MKMatthew Korban
Matthew Korban
PYPeter Youngs
Peter Youngs
SAScott T. Acton
Scott T. Acton
Deep networks are usually built first and trained later, forcing fixed capacity and wasted compute. We introduce the Autoplastic Neural Network, a training-time framework that learns structure and weights together. Layer inclusion, neuron-level connectivity, and per-layer operation mixing are jointly optimized within a fixed maximal supernetwork, whose pre-allocated components are reinforced or pruned as learning progresses; the resulting model supports per-input routing and can also be exported as a compact static architecture. A stability-gated rule coordinates architectural changes with gradient descent and yields convergence guarantees. Across vision, video, language, bioinformatics, and healthcare tasks, the approach surpasses strong static and dynamic baselines while using less computation. Unlike neural architecture search, no outer search loop is required; evolution happens inside a single learning process with cost comparable to training a single model. By unifying optimization and design, Autoplastic networks allocate capacity where needed, remove redundancy when possible, and deliver practical, resource-aware deep learning.