Papers
4
Total Citations
36
H-Index
3
About
Tae-Min Choi is a rising researcher whose work bridges two critical frontiers: time-series data imputation and neuromorphic vision for robotics. His most impactful contribution, "RDIS: Random Drop Imputation With Self-Training for Incomplete Time Series Data" (2023, 15 citations), tackles the pervasive problem of missing values in fields like healthcare and meteorology. By introducing a novel self-training framework that randomly drops observed values to create a supervised signal, Choi’s method enables models to learn robust imputation without relying on ground-truth missing data—a significant advance over traditional implicit training approaches. In parallel, his work on "In-sensor multilevel image adjustment for high-clarity contour extraction using adjustable synaptic phototransistors" (2025, 13 citations) addresses the energy and speed bottlenecks in robotic vision. By integrating data compaction directly into the sensor via adjustable synaptic phototransistors, Choi demonstrates how in-sensor processing can achieve high-clarity contour extraction without sacrificing efficiency, paving the way for low-power, real-time robotic perception. With a growing citation record and involvement in the PhaKIR 2024 challenge for surgical endoscopy analysis, Choi is establishing himself as a versatile innovator at the intersection of data science and hardware-accelerated AI.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4