Dynamic-Net: Tuning the Objective Without Re-training

11/21/2018
by   Alon Shoshan, et al.
18

One of the key ingredients for successful optimization of modern CNNs is identifying a suitable objective. To date, the objective is fixed a-priori at training time, and any variation to it requires re-training a new network. In this paper we present a first attempt at alleviating the need for re-training. Rather than fixing the network at training time, we train a "Dynamic-Net" that can be modified at inference time. Our approach considers an "objective-space" as the space of all linear combinations of two objectives, and the Dynamic-Net can traverse this objective-space at test-time, without any further training. We show that this upgrades pre-trained networks by providing an out-of-learning extension, while maintaining the performance quality. The solution we propose is fast and allows a user to interactively modify the network, in real-time, in order to obtain the result he/she desires. We show the benefits of such an approach via several different applications.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset