Fabio Pardo
Papers
2
Total Citations
14
H-Index
2
About
Fabio Pardo’s research lies at the intersection of biomechanics, reinforcement learning, and deep learning infrastructure. He is best known for **OstrichRL** (2021, 8 citations), a pioneering musculoskeletal simulation of an ostrich that enables the study of bio-mechanical locomotion. This work tackles the formidable challenge of muscle-actuated control—where bodies are overactuated and dynamics are delayed and nonlinear—bridging biomechanics, neuroscience, robotics, and graphics. Pardo’s simulation provides a rich testbed for developing and benchmarking reinforcement learning algorithms in complex, physically realistic environments. Beyond biomechanics, Pardo made a foundational contribution to deep learning portability with **Ivy** (2021, 6 citations), a templated framework that abstracts existing DL frameworks. Ivy unifies core functions across frameworks to ensure consistent call signatures, syntax, and input-output behavior, enabling researchers to write framework-agnostic code. This work addresses a critical pain point in the ML community, facilitating seamless experimentation and collaboration across TensorFlow, PyTorch, JAX, and others. Pardo’s dual focus on biologically inspired control and software infrastructure reflects a rare combination of domain science and engineering rigor. His work empowers researchers to push the boundaries of both robotic locomotion and reproducible, portable deep learning research.
Research Focus
Key Achievements
Top Papers
- 1
- 2Ivy: Templated Deep Learning for Inter-Framework Portability6 citations · 2021