hide
Free keywords:
Astrophysics, Instrumentation and Methods for Astrophysics, astro-ph.IM
Abstract:
In this paper, we present a novel artificial intelligence (AI) program that
identifies pulsars from recent surv eys using image pattern recognition with
deep neural nets---the PICS(Pulsar Image-based Classification System) AI. The
AI mimics human experts and distinguishes pulsars from noise and interferences
by looking for patterns from candidate. The information from each pulsar
candidate is synthesized in four diagnostic plots, which consist of up to
thousands pixel of image data. The AI takes these data from each candidate as
its input and uses thousands of such candidates to train its $\sim$9000
neurons. Different from other pulsar selection programs which use pre-designed
patterns, the PICS AI teaches itself the salient features of different pulsars
from a set of human-labeled candidates through machine learning. The deep
neural networks in this AI system grant it superior ability in recognizing
various types of pulsars as well as their harmonic signals. The trained AI's
performance has been validated with a large set of candidates different from
the training set. In this completely independent test, PICS ranked 264 out of
277 pulsar-related candidates, including all 56 previously known pulsars, to
the top 961 (1%) of 90008 test candidates, missing only 13 harmonics. The first
non-pulsar candidate appears at rank 187, following 45 pulsars and 141
harmonics. Performance of this system can be improved over time as more
training data are accumulated. This AI system has been integrated into the
PALFA survey pipeline and has discovered three new pulsars to date.
