skip to main content

Search for: All records

Creators/Authors contains: "Roy, J."

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. ABSTRACT Since their discovery more than 50 years ago, broad-band radio studies of pulsars have generated a wealth of information about the underlying physics of radio emission. In order to gain some further insights into this elusive emission mechanism, we performed a multifrequency study of two very well-known pulsars, PSR B0919+06 and PSR B1859+07. These pulsars show peculiar radio emission properties whereby the emission shifts to an earlier rotation phase before returning to the nominal emission phase in a few tens of pulsar rotations (also known as ‘swooshes’). We confirm the previous claim that the emission during the swoosh is not necessarily absent at low frequencies and the single pulses during a swoosh show varied behaviour at 220 MHz. We also confirm that in PSR B0919+06, the pulses during the swoosh show a chromatic dependence of the maximum offset from the normal emission phase with the offset following a consistent relationship with observing frequency. We also observe that the flux density spectrum of the radio profile during the swoosh is inverted compared to the normal emission. For PSR B1859+07, we have discovered a new mode of emission in the pulsar that is potentially quasi-periodic with a different periodicity than is seen in its swooshes. We invokemore »an emission model previously proposed in the literature and show that this simple model can explain the macroscopic observed characteristics in both pulsars. We also argue that pulsars that exhibit similar variability on short time-scales may have the same underlying emission mechanism.« less
  2. We introduce Wasserstein Adversarial Proximal Policy Optimization (WAPPO), a novel algorithm for visual transfer in Reinforcement Learning that explicitly learns to align the distributions of extracted features between a source and target task. WAPPO approximates and minimizes the Wasserstein-1 distance between the distributions of features from source and target domains via a novel Wasserstein Confusion objective. WAPPO outperforms the prior state-of-the-art in visual transfer and successfully transfers policies across Visual Cartpole and both the easy and hard settings of of 16 OpenAI Procgen environments.