Top 10 Arxiv Papers Today
2.257 Mikeys
#1. FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
Run Wang, Lei Ma, Felix Juefei-Xu, Xiaofei Xie, Jian Wang, Yang Liu
In recent years, we have witnessed the unprecedented success of generative
adversarial networks (GANs) and its variants in image synthesis. These
techniques are widely adopted in synthesizing fake faces which poses a serious
challenge to existing face recognition (FR) systems and brings potential
security threats to social networks and media as the fakes spread and fuel the
misinformation. Unfortunately, robust detectors of these AI-synthesized fake
faces are still in their infancy and are not ready to fully tackle this
emerging challenge. Currently, image forensic-based and learning-based
approaches are the two major categories of strategies in detecting fake faces.
In this work, we propose an alternative category of approaches based on
monitoring neuron behavior. The studies on neuron coverage and interactions
have successfully shown that they can be served as testing criteria for deep
learning systems, especially under the settings of being exposed to adversarial
attacks. Here, we conjecture that monitoring neuron behavior can...
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces. Run Wang, Lei Ma, Felix Juefei-Xu, Xiaofei Xie, Jian Wang, and Yang Liu https://t.co/ID9LvN9IV6
arxiv_cs_LG:
FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces. Run Wang, Lei Ma, Felix Juefei-Xu, Xiaofei Xie, Jian Wang, and Yang Liu https://t.co/CtOudYeVNk
EricSchles:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
GiorgioPatrini:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
miguelbandera:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
KouroshMeshgi:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
SingingData:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
subhobrata1:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
ajnovice:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
meverteam:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
caymanlee:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
Pol09122455:
RT @roadrunning01: FakeSpotter: A Simple Baseline for Spotting AI-Synthesized Fake Faces
pdf: https://t.co/nyrDqLtkR3
abs: https://t.co/yhM…
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 6
Total Words: 0
Unqiue Words: 0
2.249 Mikeys
#2. Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov)
Adam Hibberd, Nikolaos Perakis, Andreas M. Hein
A potential second interstellar object C/2019 Q4 (Borisov) was discovered
after the first known interstellar object 1I/'Oumuamua. Can we send a
spacecraft to this object, using existing technologies? In this paper, we
assess the technical feasibility of a mission to C/2019 Q4 (Borisov), using
existing technologies. We apply the Optimum Interplanetary Trajectory Software
(OITS) tool to generate trajectories to C/2019 Q4 (Borisov). As results, we get
the minimal DeltaV trajectory with a launch date in July 2018. For this
trajectory, a Falcon Heavy launcher could have hauled a 2 ton spacecraft to
C/2019 Q4 (Borisov). For a later launch date, results for a combined powered
Jupiter flyby with a Solar Oberth maneuver are presented. For a launch in 2030,
we could reach C/2019 Q4 (Borisov) in 2045, using the Space Launch System,
up-scaled Parker probe heatshield technology, and solid propulsion engines. A
CubeSat-class spacecraft with a mass of 3 kg could be sent to C/2019 Q4
(Borisov). If C/2019 Q4 (Borisov) turns out to be indeed an...
more |
pdf
| html
Figures
None.
Tweets
localgroupjp:
2つ目の恒星間天体候補、C/2019 Q4(Borisov)に探査機を送ろうとした場合、2018年7月に打ち上げることができていたなら、ファルコンヘビーロケットを使って2トンの宇宙船を送り込めたとのこと
https://t.co/aJ63tnNN4B
I4Interstellar:
i4is Project Lyra team publishes - Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov) - https://t.co/DPrspx3K1e https://t.co/wm59ZRGGtr
Comet2013A1:
Fresh off the arxiv #C2019Q4 #Borisov "Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov)" https://t.co/NcXlHPsCuV
StarshipBuilder:
Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov)
https://t.co/YrbQFtMaxj
scimichael:
Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov)
https://t.co/KylIVL05d7
I4Interstellar:
RT @I4Interstellar: i4is Project Lyra team publishes - Sending a Spacecraft to Interstellar Comet C/2019 Q4 (Borisov) - https://t.co/DPrspx3K1e https://t.co/wm59ZRGGtr
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 0
Unqiue Words: 0
2.195 Mikeys
#3. A Comparative Study on Transformer vs RNN in Speech Applications
Shigeki Karita, Nanxin Chen, Tomoki Hayashi, Takaaki Hori, Hirofumi Inaguma, Ziyan Jiang, Masao Someki, Nelson Enrique Yalta Soplin, Ryuichi Yamamoto, Xiaofei Wang, Shinji Watanabe, Takenori Yoshimura, Wangyou Zhang
Sequence-to-sequence models have been widely used in end-to-end speech
processing, for example, automatic speech recognition (ASR), speech translation
(ST), and text-to-speech (TTS). This paper focuses on an emergent
sequence-to-sequence model called Transformer, which achieves state-of-the-art
performance in neural machine translation and other natural language processing
applications. We undertook intensive studies in which we experimentally
compared and analyzed Transformer and conventional recurrent neural networks
(RNN) in a total of 15 ASR, one multilingual ASR, one ST, and two TTS
benchmarks. Our experiments revealed various training tips and significant
performance benefits obtained with Transformer for each task including the
surprising superiority of Transformer in 13/15 ASR benchmarks in comparison
with RNN. We are preparing to release Kaldi-style reproducible recipes using
open source and publicly available datasets for all the ASR, ST, and TTS tasks
for the community to succeed our exciting outcomes.
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
A Comparative Study on Transformer vs RNN in Speech Applications. Karita, Chen, Hayashi, Hori, Inaguma, Jiang, Someki, Soplin, Yamamoto, Wang, Watanabe, Yoshimura, and Zhang https://t.co/ZG2tuYIGSz
kari_tech:
https://t.co/3QPtNiEk46 Our preprint "A Comparative Study on Transformer vs RNN in Speech Applications" (ASRU2019) is available! https://t.co/g6XFlxpYYB
ballforest:
RT @kari_tech: https://t.co/3QPtNiEk46 Our preprint "A Comparative Study on Transformer vs RNN in Speech Applications" (ASRU2019) is availa…
kastnerkyle:
RT @kari_tech: https://t.co/3QPtNiEk46 Our preprint "A Comparative Study on Transformer vs RNN in Speech Applications" (ASRU2019) is availa…
ymas0315:
RT @kari_tech: https://t.co/3QPtNiEk46 Our preprint "A Comparative Study on Transformer vs RNN in Speech Applications" (ASRU2019) is availa…
r9y9:
RT @kari_tech: https://t.co/3QPtNiEk46 Our preprint "A Comparative Study on Transformer vs RNN in Speech Applications" (ASRU2019) is availa…
chbalajitilak:
RT @arxiv_cscl: A Comparative Study on Transformer vs RNN in Speech Applications https://t.co/KDiMBh0O1D
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 13
Total Words: 0
Unqiue Words: 0
2.174 Mikeys
#4. White-Box Adversarial Defense via Self-Supervised Data Estimation
Zudi Lin, Hanspeter Pfister, Ziming Zhang
In this paper, we study the problem of how to defend classifiers against
adversarial attacks that fool the classifiers using subtly modified input data.
In contrast to previous works, here we focus on the white-box adversarial
defense where the attackers are granted full access to not only the classifiers
but also defenders to produce as strong attacks as possible. In such a context
we propose viewing a defender as a functional, a higher-order function that
takes functions as its argument to represent a function space, rather than
fixed functions conventionally. From this perspective, a defender should be
realized and optimized individually for each adversarial input. To this end, we
propose RIDE, an efficient and provably convergent self-supervised learning
algorithm for individual data estimation to protect the predictions from
adversarial attacks. We demonstrate the significant improvement of adversarial
defense performance on image recognition, eg, 98%, 76%, 43% test accuracy on
MNIST, CIFAR-10, and ImageNet datasets...
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
White-Box Adversarial Defense via Self-Supervised Data Estimation. Zudi Lin, Hanspeter Pfister, and Ziming Zhang https://t.co/w0cY6YXt0h
arxivml:
"White-Box Adversarial Defense via Self-Supervised Data Estimation",
Zudi Lin, Hanspeter Pfister, Ziming Zhang
https://t.co/YWbgzEnunU
arxiv_cs_LG:
White-Box Adversarial Defense via Self-Supervised Data Estimation. Zudi Lin, Hanspeter Pfister, and Ziming Zhang https://t.co/Fr3spOyQWg
arxiv_cscv:
White-Box Adversarial Defense via Self-Supervised Data Estimation https://t.co/sRy0ug3l3f
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 0
Unqiue Words: 0
2.173 Mikeys
#5. Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering
Shiyue Zhang, Mohit Bansal
Text-based Question Generation (QG) aims at generating natural and relevant
questions that can be answered by a given answer in some context. Existing QG
models suffer from a "semantic drift" problem, i.e., the semantics of the
model-generated question drifts away from the given context and answer. In this
paper, we first propose two semantics-enhanced rewards obtained from downstream
question paraphrasing and question answering tasks to regularize the QG model
to generate semantically valid questions. Second, since the traditional
evaluation metrics (e.g., BLEU) often fall short in evaluating the quality of
generated questions, we propose a QA-based evaluation method which measures the
QG model's ability to mimic human annotators in generating QA training data.
Experiments show that our method achieves the new state-of-the-art performance
w.r.t. traditional metrics, and also performs best on our QA-based evaluation
metrics. Further, we investigate how to use our QG model to augment QA datasets
and enable semi-supervised QA. We...
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering. Shiyue Zhang and Mohit Bansal https://t.co/JAV8U4ybe5
arxivml:
"Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering",
Shiyue Zhang, Mohit Bans…
https://t.co/XHCrUAdEIQ
arxiv_cs_LG:
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering. Shiyue Zhang and Mohit Bansal https://t.co/wVa2BULeLu
arxiv_cscl:
Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering https://t.co/LQ3AvzwQWg
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 2
Total Words: 0
Unqiue Words: 0
2.144 Mikeys
#6. Five misconceptions about black holes
Jorge Pinochet
Given the great interest that black holes arouse among non-specialists, it is
important to analyse misconceptions related to them. According to the author,
the most common misconceptions are that: (1) black holes are formed from
stellar collapse; (2) they are very massive; (3) they are very dense; (4) their
gravity absorbs everything; and (5) they are black. The objective of this work
is to analyse and correct these misconceptions. This article may be useful as
pedagogical material in high school physics courses or in introductory courses
in undergraduate physics.
more |
pdf
| html
Figures
None.
Tweets
emulenews:
Five misconceptions about black holes, by Jorge Pinochet in Physics Education (2019)
https://t.co/KAOOrlR1gM https://t.co/KMnb0JhxYL
LCTTA:
Five misconceptions about black holes. (arXiv:1909.06006v1 [physics.pop-ph]) https://t.co/GMPdAvZWt2
DrRaulTorres:
RT @emulenews: Five misconceptions about black holes, by Jorge Pinochet in Physics Education (2019)
https://t.co/KAOOrlR1gM https://t.co/KM…
kur41:
RT @emulenews: Five misconceptions about black holes, by Jorge Pinochet in Physics Education (2019)
https://t.co/KAOOrlR1gM https://t.co/KM…
cuantizado:
RT @emulenews: Five misconceptions about black holes, by Jorge Pinochet in Physics Education (2019)
https://t.co/KAOOrlR1gM https://t.co/KM…
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 1
Total Words: 0
Unqiue Words: 0
2.13 Mikeys
#7. Reinforcement Learning: a Comparison of UCB Versus Alternative Adaptive Policies
Wesley Cowan, Michael N. Katehakis, Daniel Pirutinsky
In this paper we consider the basic version of Reinforcement Learning (RL)
that involves computing optimal data driven (adaptive) policies for Markovian
decision process with unknown transition probabilities. We provide a brief
survey of the state of the art of the area and we compare the performance of
the classic UCB policy of \cc{bkmdp97} with a new policy developed herein which
we call MDP-Deterministic Minimum Empirical Divergence (MDP-DMED), and a method
based on Posterior sampling (MDP-PS).
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
Reinforcement Learning: a Comparison of UCB Versus Alternative Adaptive Policies. Wesley Cowan, Michael N. Katehakis, and Daniel Pirutinsky https://t.co/mSvyz7Xk4z
arxivml:
"Reinforcement Learning: a Comparison of UCB Versus Alternative Adaptive Policies",
Wesley Cowan, Michael N． Kateha…
https://t.co/Qm05S3DRiT
arxiv_cs_LG:
Reinforcement Learning: a Comparison of UCB Versus Alternative Adaptive Policies. Wesley Cowan, Michael N. Katehakis, and Daniel Pirutinsky https://t.co/56wEYBpP1x
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 0
Unqiue Words: 0
2.13 Mikeys
#8. Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs
Jonas Kubilius, Martin Schrimpf, Ha Hong, Najib J. Majaj, Rishi Rajalingham, Elias B. Issa, Kohitij Kar, Pouya Bashivan, Jonathan Prescott-Roy, Kailyn Schmidt, Aran Nayebi, Daniel Bear, Daniel L. K. Yamins, James J. DiCarlo
Deep convolutional artificial neural networks (ANNs) are the leading class of
candidate models of the mechanisms of visual processing in the primate ventral
stream. While initially inspired by brain anatomy, over the past years, these
ANNs have evolved from a simple eight-layer architecture in AlexNet to
extremely deep and branching architectures, demonstrating increasingly better
object categorization performance, yet bringing into question how brain-like
they still are. In particular, typical deep models from the machine learning
community are often hard to map onto the brain's anatomy due to their vast
number of layers and missing biologically-important connections, such as
recurrence. Here we demonstrate that better anatomical alignment to the brain
and high performance on machine learning as well as neuroscience measures do
not have to be in contradiction. We developed CORnet-S, a shallow ANN with four
anatomically mapped areas and recurrent connectivity, guided by Brain-Score, a
new large-scale composite of neural and...
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs. Kubilius, Schrimpf, Hong, Majaj, Rajalingham, Issa, Kar, Bashivan, Prescott-Roy, Schmidt, Nayebi, Bear, Yamins, and DiCarlo https://t.co/NmuEKcb1WK
arxivml:
"Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs",
Jonas Kubilius, Martin Schrimpf, Ha Ho…
https://t.co/ASnLLt2E3x
arxiv_cs_LG:
Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs. Kubilius, Schrimpf, Hong, Majaj, Rajalingham, Issa, Kar, Bashivan, Prescott-Roy, Schmidt, Nayebi, Bear, Yamins, and DiCarlo https://t.co/WodOXHvfL1
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 14
Total Words: 0
Unqiue Words: 0
2.13 Mikeys
#9. Deep Adversarial Belief Networks
Yuming Huang, Ashkan Panahi, Hamid Krim, Yiyi Yu, Spencer L. Smith
We present a novel adversarial framework for training deep belief networks
(DBNs), which includes replacing the generator network in the methodology of
generative adversarial networks (GANs) with a DBN and developing a highly
parallelizable numerical algorithm for training the resulting architecture in a
stochastic manner. Unlike the existing techniques, this framework can be
applied to the most general form of DBNs with no requirement for back
propagation. As such, it lays a new foundation for developing DBNs on a par
with GANs with various regularization units, such as pooling and normalization.
Foregoing back-propagation, our framework also exhibits superior scalability as
compared to other DBN and GAN learning techniques. We present a number of
numerical experiments in computer vision as well as neurosciences to illustrate
the main advantages of our approach.
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
Deep Adversarial Belief Networks. Yuming Huang, Ashkan Panahi, Hamid Krim, Yiyi Yu, and Spencer L. Smith https://t.co/CJSgXdpYvj
arxivml:
"Deep Adversarial Belief Networks",
Yuming Huang, Ashkan Panahi, Hamid Krim, Yiyi Yu, Spencer L． Smith
https://t.co/PL6FesaG55
arxiv_cs_LG:
Deep Adversarial Belief Networks. Yuming Huang, Ashkan Panahi, Hamid Krim, Yiyi Yu, and Spencer L. Smith https://t.co/H0cMdWAStA
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 5
Total Words: 0
Unqiue Words: 0
2.13 Mikeys
#10. Deep Learned Path Planning via Randomized Reward-Linked-Goals and Potential Space Applications
Tamir Blum, William Jones, Kazuya Yoshida
Space exploration missions have seen use of increasingly sophisticated
robotic systems with ever more autonomy. Deep learning promises to take this
even a step further, and has applications for high-level tasks, like path
planning, as well as low-level tasks, like motion control, which are critical
components for mission efficiency and success. Using deep reinforcement
end-to-end learning with randomized reward function parameters during training,
we teach a simulated 8 degree-of-freedom quadruped ant-like robot to travel
anywhere within a perimeter, conducting path plan and motion control on a
single neural network, without any system model or prior knowledge of the
terrain or environment. Our approach also allows for user specified waypoints,
which could translate well to either fully autonomous or
semi-autonomous/teleoperated space applications that encounter delay times. We
trained the agent using randomly generated waypoints linked to the reward
function and passed waypoint coordinates as inputs to the neural network....
more |
pdf
| html
Figures
None.
Tweets
BrundageBot:
Deep Learned Path Planning via Randomized Reward-Linked-Goals and Potential Space Applications. Tamir Blum, William Jones, and Kazuya Yoshida https://t.co/Lb1qLiL7rX
arxivml:
"Deep Learned Path Planning via Randomized Reward-Linked-Goals and Potential Space Applications",
Tamir Blum, Willi…
https://t.co/hThfd1vPJa
arxiv_cs_LG:
Deep Learned Path Planning via Randomized Reward-Linked-Goals and Potential Space Applications. Tamir Blum, William Jones, and Kazuya Yoshida https://t.co/9hmTc1oOR9
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 0
Unqiue Words: 0
About
Assert is a website where the best academic papers on arXiv (computer science, math, physics), bioRxiv (biology), BITSS (reproducibility), EarthArXiv (earth science), engrXiv (engineering), LawArXiv (law), PsyArXiv (psychology), SocArXiv (social science), and SportRxiv (sport research) bubble to the top each day.
Papers are scored (in real-time) based on how verifiable they are (as determined by their Github repos) and how interesting they are (based on Twitter).
To see top papers, follow us on twitter @assertpub_ (arXiv), @assert_pub (bioRxiv), and @assertpub_dev (everything else).
To see beautiful figures extracted from papers, follow us on Instagram.
Tracking 189,566 papers.
Search
Sort results based on if they are interesting or reproducible.
Interesting
Reproducible