Frank Liu

About

Frank Liu is a researcher focused on advancing continual learning for dynamic, real-world AI systems such as self-driving vehicles, surveillance drones, and robotics. His most cited work, "Single-Net Continual Learning with Progressive Segmented Training" (2019, 13 citations), introduces a novel framework that enables a single neural network to sequentially learn new tasks without catastrophic forgetting—a critical challenge in lifelong machine learning. By segmenting the network’s parameters and progressively training them, Liu’s approach preserves previously acquired knowledge while adapting to new data streams, offering a more efficient and scalable alternative to multi-network ensembles. This contribution has garnered attention for its practical applicability in resource-constrained environments where model size and computational cost are limited. Liu’s research directly addresses the growing need for systems that can continuously learn from evolving streams of information, making his work foundational for deploying AI in autonomous and adaptive technologies. His achievements highlight a commitment to bridging the gap between theoretical continual learning and real-world deployment.

Research Focus

Key Achievements

Top Papers

1. 1
   Single-Net Continual Learning with Progressive Segmented Training

   13 citations · 2019
2. 2
   Single-Net Continual Learning with Progressive Segmented Training (PST)

   4 citations · 2019

Key Collaborators

Contact & Links