Felipe Vieira Frujeri
Papers
1
Total Citations
14
H-Index
1
About
Felipe Vieira Frujeri is a researcher at the forefront of robotics and machine learning, with a focus on developing scalable, data-driven approaches to autonomous systems. His key research areas include robotic perception, action planning, and causal representation learning, where he seeks to reduce the need for handcrafted engineering in complex robotic architectures. Frujeri’s major contribution is the introduction of the Perception-Action Causal Transformer (PACT), a novel paradigm for autoregressive robotics pre-training. This work, published in 2023, draws inspiration from large language models to create a unified framework that learns causal relationships between sensory inputs and motor actions, enabling robots to generalize across tasks without task-specific programming. With 14 citations to date, PACT has quickly gained attention for its potential to streamline robotics development. Frujeri’s work stands out for bridging the gap between causal inference and transformer-based learning, offering a path toward more adaptable and intelligent robotic systems. His research is particularly valuable for students and engineers seeking to move beyond traditional modular designs toward end-to-end learning in robotics.
Research Focus
Key Achievements
Top Papers
- 1