Although an explanation of a single prediction provides some understanding into the reliability of the classifier to the user, it is not sufficient to evaluate and assess trust in the model as a whole. We propose to give a global understanding of the model by explaining a set of individual in...
In this essay, I review the concept of color-blind racism and its application to the Obama phenomenon. I also revisit some of my past predictions for Obama's presidency and evaluate their accuracy halfway through his term. Finally, I offer suggestions for constructing a genuine social movement...
Flew argued that a similar burden of proof applies to propositions about God. In his analogy, asserting that God exists is like asserting that a crime occurred and that the burden of proof lies on the person making the assertion. It is unreasonable, he goes on, to expect the other side t...
Likewise for image classification, an in- terpretable representation may be a binary vector indicating the “presence” or “absence” of a contiguous patch of similar pixels (a super-pixel), while the classifier may represent the image as a tensor with three color channels per pixel. We deno...
First, by having a lower academic standard for individuals that share something in common (i.e. their skin color or their heritage), schools actively promote the belief that, on average, individuals who share that skin color or heritage are less skilled and less qualified. Say, for instance,...
embeddings. Likewise for image classification, an in- terpretable representation may be a binary vector indicating the “presence” or “absence” of a contiguous patch of similar pixels (a super-pixel), while the classifier may represent the image as a tensor with three color channels per ...