Page 143 - Rapid Learning in Robotics

9.2 “Investment Learning” or “Mixture-of-Expertise” Architecture 129

Input

Context Gating
Network

T-Box Σ
Task Variables Expert 2 Output
Expert 1

T-Box

T-Box
Expert 3

T-Box
Expert N ‘‘Mixture-of-Experts’’

‘‘Mixture-of-Expertise’’
Input

Context
Meta
Network Parameters
ω

Task Variables
T-Box
Expert

Output

Figure 9.4: The “Mixture-of-Experts” architecture versus the “Mixture-of-
Expertise” architecture.