QAOA Suffers from Barren Plateaus for Most MaxCut Instances
Analysis
Key Takeaways
- •QAOA suffers from barren plateaus for most MaxCut instances, making training difficult.
- •The DLA dimension grows exponentially for a large fraction of graphs.
- •A new algorithm is developed to analyze the DLA, improving computational efficiency.
- •The findings suggest limitations in QAOA's practical applicability for MaxCut.
“The paper shows that the DLA dimension grows as $Θ(4^n)$ for weighted graphs (with continuous weight distributions) and almost all unweighted graphs, implying barren plateaus.”