Max Robotics — Global Service Robot Directory + US Certification

About

Jonathan Ho is a versatile machine learning and robotics researcher whose work spans trajectory optimization, imitation learning, and deep generative modeling. He first gained recognition through his foundational contributions to robotic motion planning, co-developing Sequential Convex Optimization (SCO) approaches that enabled efficient, collision-free trajectory generation from simple initializations — work that has accumulated over 840 and 429 citations respectively, establishing him as a key figure in optimization-based planning. Ho then broadened his scope into learning-based robotics, contributing influential work on one-shot imitation learning (229 citations), which demonstrated how robots could generalize tasks from minimal demonstrations — a significant step toward more adaptable autonomous systems. His 2016 work on non-rigid registration for learning from demonstrations further enriched this portfolio. Expanding into deep learning architectures, Ho co-developed Axial Transformers (364 citations), an efficient self-attention mechanism for high-dimensional data that influenced subsequent generative modeling research. His meta-learning work on shared hierarchical policies (117 citations) reflects a consistent interest in sample efficiency and generalization across tasks. Taken together, Ho's research represents a compelling trajectory from classical optimization to modern deep learning, with meaningful impact across robotics and machine learning communities.