We introduce the “Energy-based Generative Adversarial Network” model (EBGAN) which views the discriminator as an energy function that associates low energies with the regions near the data manifold and higher energies with other regions. Similar to the probabilistic GANs, a generator is trained to produce contrastive samples with minimal energies, while the discriminator is trained to assign hi...

Generative Adversarial Networks (GANs) have recently been proposed as a promising avenue towards learning generative models with deep neural networks. While GANs have demonstrated state-of-the-art performance on multiple vision tasks, their learning dynamics are not yet well understood, both in theory and in practice. To address this issue, we take a first step towards a rigorous study of GAN d...

The performance of a conventional intensity modulation direct detection analog optical links is limited by the high noise power associated with large optical carrier power. This optical carrier can be suppressed by using optical frequency modulation in conjunction with a frequency discriminator. In this paper, we propose a design of an integrated optical frequency discriminator, based on optica...

A simple amplitude discriminator with two ways of triggering DC-level monitoring is described. This system is based on easily available inexpensive integrated circuits.

For f(X) ∈ Z[X], let Df (n) be the least positive integer k for which f(1), . . . , f(n) are distinct modulo k. Several results have been proven about the function Df in recent years, culminating in Moree’s characterization of Df (n) whenever f lies in a certain (large) subset of Z[X] and n is sufficiently large. We give several improvements of Moree’s result, as well as further results on the ...

Discriminator architecture is shown in Figure 1 . In our setup the job of the discriminator is to analyze the local statistics of images. Therefore, after five convolutional layers with occasional stride we perform global average pooling. The result is processed by two fully connected layers, followed by a 2-way softmax. We perform 50% dropout after the global average pooling layer and the firs...

This paper focuses on low cost production testing of the far-out phase noise of PLL ICs using the delay line discriminator method. It describes two different delay line discriminator (DLD) implementations for phase noise measurements at large frequency offsets from the carrier. The calibration method using an FM calibration signal is described in detail, both mathematically and graphically. The...