inacionline
LightNews LightNews
banner
inacionline.bsky.social
inacionline
@inacionline.bsky.social
82 followers 170 following 90 posts
informatics engineer;
explainability and meta-learning research;
Posts Replies Media Videos
inacionline.bsky.social
i mean, look
Figure 1: How the path kernel measures similarity between examples. In this two-dimensional illustration, as the weights follow a path on the plane during training, the model’s gradients (vectors on the weight plane) for x, x1 and x2 vary along it. The kernel K(x, x1) is the integral of the dot product of the gradients ∇wy(x) and ∇wy(x1) over the path, and similarly for K(x, x2). Because on average over the weight path ∇wy(x) · ∇wy(x1) is greater than ∇wy(x) · ∇wy(x2), y1 has more influence than y2 in predicting y, all else being equal.
March 4, 2025 at 9:50 PM
inacionline.bsky.social
as big as it needed to be (maybe too big at times)
the map of Final Fantasy I
March 4, 2025 at 9:32 PM
inacionline.bsky.social
some real wonders out there
July 11, 2023 at 6:48 PM