Company: Deloitte USI_5sep
Difficulty: easy
Two models are evaluated on the same dataset. Model A has very low training error but much higher validation error, while Model B has slightly higher training error but lower validation error. Which model is likely to generalize better?