Papers
4
Total Citations
38
H-Index
4
About
Minju Jung is a robotics and cognitive systems researcher whose work sits at the intersection of deep learning, computational neuroscience, and humanoid robot cognition. Her research focuses on developing biologically inspired neural network architectures that enable robots to acquire and integrate complex cognitive skills — including visual recognition, attention switching, action generation, and planning — through experience-based learning rather than explicit programming. Jung's most recognized contribution is the Visuo-Motor Deep Dynamic Neural Network (VMDNN), introduced in her 2015 work (13 citations), which demonstrated how coordinated cognitive behavior in humanoid robots can emerge from learned visuo-motor-attentional dynamics. Her subsequent 2016 study (10 citations) extended this framework toward seamless cognitive integration, while her 2018 research explored predictive coding-type recurrent neural networks for goal-directed visuomotor planning in robotic arms. Most ambitiously, her 2021 paper (10 citations) tackled the challenge of content-agnostic generalization, drawing on active inference, working memory, and planning to investigate how robots might manipulate unfamiliar objects — mirroring flexible human cognition. Across her career, Jung has consistently pushed toward robots that learn, adapt, and reason more like biological minds, making her work highly relevant to researchers in embodied AI and cognitive robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4