Rewards are scores that are computed on generated molecules and sent back to the generator, which learns from these scores what did and didn't work, and tries to generate new molecules with better scores using a method known as Reinforcement Learning.
In Makya, users are not restricted in the number of rewards they can select to guide their generators.
However,
The more rewards the user selects, the more generation is constrained, and the less the generator can explore the available chemical space in search of new, interesting molecules.
If your generator fails at optimizing all the rewards that were selected, it might be too constrained: try starting a new generation with only the most important rewards. Limiting the number of rewards you select prevents diluting the goals that really are the most important to your project!