The more accurate a face recognition model becomes, the less it sees faces like a human being.
There is a direct trade-off between how well a model identifies a person and how much its perception aligns with human eyes. To achieve superhuman accuracy, these models abandon the perceptual logic that we use to recognize friends and family. They focus on mathematical features that are invisible or irrelevant to the human brain. This means that as AI gets better, it also becomes more alien and harder for us to understand. We cannot assume that better AI means AI that thinks more like we do.
Trade-off between performance and human-like perception in face recognition models
PsyArXiv · bz3f9_v3
Deep learning models for face recognition are widely adopted in cognitive neuroscience, yet it is not well understood whether their excellent performance means they “see” faces like humans. Here, we collected a relatively large dataset of human face-similarity judgments to assess how well prominent face recognition models align with human perception. We found that models with high — but not the highest — recognition performance are often best aligned with human similarity judgments. We further t