Lior Wolf, Tomer Galanti
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
In the context of learning to map an input I to a function h I : X → R, two alternative methods are compared: (i) an embedding-based method, which learns a fixed function in which I is encoded as a conditioning signal e(I) and the learned function takes the form h I (x) = q(x, e(I)), and (ii) hypernetworks, in which the weights θ I of the function h I (x) = g(x; θ I ) are given by a hypernetwork f as θ I = f (I). In this paper, we define the property of modularity as the ability to effectively learn a different function for each input instance I. For this purpose, we adopt an expressivity perspective of this property and extend the theory of [10] and provide a lower bound on the complexity (number of trainable parameters) of neural networks as function approximators, by eliminating the requirements for the approximation method to be robust. Our results are then used to compare the complexities of q and g, showing that under certain conditions and when letting the functions e and f be as large as we wish, g can be smaller than q by orders of magnitude. This sheds light on the modularity of hypernetworks in comparison with the embedding-based method. Besides, we show that for a structured target function, the overall number of trainable parameters in a hypernetwork is smaller by orders of magnitude than the number of trainable parameters of a standard neural network and an embedding method.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2002.10006")
get_code_for_paper("2002.10006")
have("2002.10006")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2002.10006.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2002.10006)
Connect an agent — have() is free.