A multi-dimensional view of the ML researcher

Status Quo: A one-dimensional view

• We tend to have a one-dimensional view on ourselves and other ML researchers

• For example, I often hear sentences like “X is amazing”

• However, it is extremely rare that someone is truly great in all aspects of research

• In this blog post, I will attempt to explain why we have such a shallow view on ourselves and others

• Further, I will argue for why it is time to adopt a more differentiated perspective

Cognitive biases: The Halo effect

Expectations: We ought to be good at everything

• While we are specialists in our research field, we are often expected to be generalists when it comes to our research skill set

• In reality, we are of course much more complex individuals

• We often take on very different roles, and each of those roles requires a different set of skills

Remedies: More differentiated feedback

Why all of this matters: Building teams