[BUG] In Choosing the goal for the next round of experiments

[SwordFaith]

I guess greedily maximize validation error should be minimize validation error.

Example goals include:

• Try a potential improvement to the pipeline (e.g. a new regularizer, preprocessing choice, etc.).
• Understand the impact of a particular model hyperparameter (e.g. the activation function)
• Greedily maximize validation error.