SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2304.00898 · CVPR · 2023

Tunable Convolutions with Parametric Multi-Loss Optimization

Steven Mcdonagh, Aleš Leonardis, Matteo Maggioni, Francesca Babiloni, Thomas Tanay

arXiv · PDF · Open in the Atlas

Code that ran

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.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Behavior of neural networks is irremediably determined by the specific loss and data used during training. However it is often desirable to tune the model at inference time based on external factors such as preferences of the user or dynamic characteristics of the data. This is especially important to balance the perception-distortion tradeoff of ill-posed image-to-image translation tasks. In this work, we propose to optimize a parametric tunable convolutional layer, which includes a number of different kernels, using a parametric multi-loss, which includes an equal number of objectives. Our key insight is to use a shared set of parameters to dynamically interpolate both the objectives and the kernels. During training, these parameters are sampled at random to explicitly optimize all possible combinations of objectives and consequently disentangle their effect into the corresponding kernels. During inference, these parameters become interactive inputs of the model hence enabling reliable and consistent control over the model behavior. Extensive experimental results demonstrate that our tunable convolutions effectively work as a drop-in replacement for traditional convolutions in existing neural networks at virtually no extra computational cost, outperforming state-of-the-art control strategies in a wide range of applications; including image denoising, deblurring, super-resolution, and style transfer.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2304.00898")
get_code_for_paper("2304.00898")
have("2304.00898")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2304.00898.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2304.00898.svg)](https://syntology.ai/paper/2304.00898)

Connect an agent — have() is free.