Students and newcomers · 8-12 weeks
Beginner Path
Build the vocabulary and technical foundations needed to understand Physical AI research.
Direct answer
Who is the Beginner Path for?
Beginner Path is designed for students and newcomers and is structured as a 8-12 weeks sequence.
Course sequence
Complete the stages in order. Each page includes definitions, prerequisites, paper explainers, and recommended next steps.
- 01Open →
Foundations
The mathematics, programming, machine learning, vision, and robotics concepts needed to study Physical AI.
- 02Open →
Physical AI
AI systems that perceive, reason, and act through physical machines in the real world.
- 03Open →
Embodied Intelligence
Intelligence that emerges through an agent's body, sensors, actions, and interaction with an environment.
- 04Open →
Computer Vision
Methods that let machines extract useful structure from images and video.
- 05Open →
Egocentric Data
First-person video, audio, gaze, hand-object interaction, and wearable sensor data used to teach AI what humans see and do.
- 06Open →
Robot Learning
Methods that allow robots to acquire behavior from demonstrations, rewards, interaction, or generated experience.
- 07Open →
World Models
Learned models that predict environment dynamics and possible future outcomes.