Top 10 Arxiv Papers Today in Computer Science


2.277 Mikeys
#1. Gradient Descent Finds Global Minima of Deep Neural Networks
Simon S. Du, Jason D. Lee, Haochuan Li, Liwei Wang, Xiyu Zhai
Gradient descent finds a global minimum in training deep neural networks despite the objective function being non-convex. The current paper proves gradient descent achieves zero training loss in polynomial time for a deep over-parameterized neural network with residual connections (ResNet). Our analysis relies on the particular structure of the Gram matrix induced by the neural network architecture. This structure allows us to show the Gram matrix is stable throughout the training process and this stability implies the global optimality of the gradient descent algorithm. Our bounds also shed light on the advantage of using ResNet over the fully connected feedforward architecture; our bound requires the number of neurons per layer scaling exponentially with depth for feedforward networks whereas for ResNet the bound only requires the number of neurons per layer scaling polynomially with depth. We further extend our analysis to deep residual convolutional neural networks and obtain a similar convergence result.
more | pdf | html
Figures
None.
Tweets
newsyc20: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/Fmcc3SDq7G (https://t.co/IGaOH0eR1D)
HNTweets: Gradient Descent Finds Global Minima of Deep Neural Networks: https://t.co/Lau0L7dys7 Comments: https://t.co/gJcjJldJ3i
deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
newsyc100: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/qi3pM5oab3 (https://t.co/ooQxRaAHap)
newsyc50: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/wk9NQBSFm2 (https://t.co/l2OxNauBeG)
roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
WAWilsonIV: Some suitable generalization of this paper could shed light on a question I sometimes think about: "why do neural networks even work at all?" https://t.co/owrB5Fnv0h
BrundageBot: Gradient Descent Finds Global Minima of Deep Neural Networks. Simon S. Du, Jason D. Lee, Haochuan Li, Liwei Wang, and Xiyu Zhai https://t.co/YcCc6xgN19
LukeSpear: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/EYvZQqcnFk #hn #startups #privacy
Synced_Global: "Gradient Descent Finds Global Minima of Deep Neural Networks" by researchers from @CarnegieMellon, @PKU1898, @USC and @MITEECS Read the full paper at https://t.co/huMYfqr1ys https://t.co/2jJzQS6qxi
loretoparisi: #GradientDescend finds a global minimum in training deep neural networks despite the objective function being non-convex proving gradient achieves zero training loss in polynomial time #DeepLearning https://t.co/soJnJwi953
harisamin: WTF is this even possible https://t.co/CdahPoOu28 ? I haven't read the paper yet, wondering what your thoughts are @spsaaibi
KloudStrife: 'Gradient Descent finds global minima of deep NNs'. Applies to resnets, in the convolutional case, no unrealistic assumptions. Important paper if confirmed. https://t.co/6soCfAPnJU
hn_frontpage: Gradient Descent Finds Global Minima of Deep Neural Networks L: https://t.co/dfFLtmVs2B C: https://t.co/RVx6tBYXCN
abhatt2: Du et al. (https://t.co/Uuygm49DYr) and Allen-Zhu et al. (https://t.co/EAnU5M7Bqg) independently seem to have solved a basic theory problem in modern ML: efficient convergence of over-parameterized deep neural nets. Very clean theorems, not many assumptions. Pretty impressive!
tisamit: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/OAsXOKgFEj #datascience #analytics
InnoveoPartners: Interesting by AI ? You have to read this "Gradient Descent Finds Global Minima of Deep Neural Networks" https://t.co/5oCVb1WCan https://t.co/lCWCQcYW69
_bha1: https://t.co/s8Ojo0Awc9 Gradient Descent Finds Global Minima of Deep Neural Networks. https://t.co/13ByZqrRoX
gmwagner: I've found ResNets and ConNN have similar performance on the same data, after reading this paper I might need to investigate that more. https://t.co/B5zuGAQzqj
arxivml: "Gradient Descent Finds Global Minima of Deep Neural Networks", Simon S. Du, Jason D. Lee, Haochuan Li, Liwei Wang,… https://t.co/231Dzu6qd6
reddit_ml: [1811.03804] Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/NlFOuwmWJj
nmfeeds: [AI] https://t.co/7qt13PxoCQ Gradient Descent Finds Global Minima of Deep Neural Networks. Gradient descent finds a global...
nmfeeds: [CV] https://t.co/7qt13PxoCQ Gradient Descent Finds Global Minima of Deep Neural Networks. Gradient descent finds a global...
nmfeeds: [O] https://t.co/7qt13PxoCQ Gradient Descent Finds Global Minima of Deep Neural Networks. Gradient descent finds a global ...
hereticreader: [1811.03804] Gradient Descent Finds Global Minima of Deep Neural Networks - https://t.co/9D1COPFWPY https://t.co/XPxp4cTzkN
DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
SciFi: Gradient Descent Finds Global Minima of Deep Neural Networks. https://t.co/MjmXyUF2cq
angsuman: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/3OZMkmEbkC
betterhn50: 55 – Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/jX9DHfr6Jm
DavidpichKh: "Gradient Descent Finds Global Minima of Deep Neural Networks" Du et al.: https://t.co/smfmtzMrrT #DeepLearning https://t.co/6qO2A0dPqr
arxiv_cscv: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/D6u9mSXJwL
arxiv_cscv: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/D6u9mSG88b
doctorSturza: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/VFizAyqXV8
hackernewsfeed: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/kmqjVk6zRg
yuxili99: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/VlPF18nUjK
hackernews100: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/EqvbM5Dkjp
hackernewsrobot: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/N8adVA0lUk
betterhn100: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/5zgLBUv7DO
hardmaru: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
rasbt: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
hugo_larochelle: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
rabkhan25: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/znqOuUM4kp
KyleCranmer: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
botnet_hunter: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
cosnet_bifi: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
ParisMLgroup: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
PaaSDev: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
tyrell_turing: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
PRONOjits: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
chrshmmmr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
hmCuesta: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
RaZ0R3: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
jaialkdanel: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
tjmlab: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
__DaLong: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
aasensior: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
nsdual: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MaxALittle: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
terashimahiroki: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
GonzaloBarria: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
ialuronico: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
stats385: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
emilioleton: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
AmineKorchiMD: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
SeguiSanti: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
xiangrenUSC: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
DrZeeshanZia: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
sp4ghet: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
datanerdword: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
harshkn: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
adelong: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
PatrickOmid: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
gregvidy: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Klevis_Ramo: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
serrjoa: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
JavierBurroni: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
arm_gilles: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
daubman: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
onsen_zuki: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Corbera_Sergio: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
fmailhot: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
thepersonwithin: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
manuelbaltieri: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
abojchevski: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
WeisiG: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
ukhndlwl: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
tkchaki: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
mattpetersen_ai: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
arora_manuel: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
RemiCadene: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
one_twit_wonder: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
stats285: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
honasu: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
kjgeras: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
letranger14: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
evansdianga: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
karthiknrao: RT @reddit_ml: [1811.03804] Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/NlFOuwmWJj
ernire: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Tanaygahlot: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
HaithamKhedr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
arthpajot: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
billderose: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MattScicluna: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
random_agent: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
sahilsingla47: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MatteoMasperoNL: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
1sdom: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
alx_eco: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
95Rohan: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MarcoZorzi: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
necoleman: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
AssistedEvolve: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
l_kiraly: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MohanraamS: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
aminedotin: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
mizvladimir: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
jlhrzn: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
RndWalk: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
siraferradans: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
quidpr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
michal_sustr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Jsevillamol: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MSerhanCan: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
namhoonlee09: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
nikos3388: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
marsusensei: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
abhijithtn: RT @deeplearning4j: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/ABH4vz2GnZ #deeplearning #machinelearning
GaryTheGammarid: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
i_shikhar98: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Saeed_KH_: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
feiwang03: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
_reachsumit: RT @hn_frontpage: Gradient Descent Finds Global Minima of Deep Neural Networks L: https://t.co/dfFLtmVs2B C: https://t.co/RVx6tBYXCN
coreyamyers: RT @DataSciFact: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/9VbSMulFLw
gevero: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
allahwala08: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
CLagares7: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
ociule: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
mnrmja007: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
adn_twitts: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
PeterMitrano: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
osamaadelshokry: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
tttorrr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
WeikangGong: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
hai_t_pham: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MozejkoMarcin: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
kli_nlpr: RT @yuxili99: Gradient Descent Finds Global Minima of Deep Neural Networks https://t.co/VlPF18nUjK
psmrustham: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Mazen_Ezzeddine: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
mattbeach42: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
jeandut14000: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
yumakajihara: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
gandhikanishk: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
mzadrogaPL: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
cezzo_sw: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Epsilon_Lee: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
vishnu_lsvsr: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
vi_shall_c: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Tsingggg: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
thinkmachine1: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
do_dreamo: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
yarphs: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
nitishjoshi23: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
pcp_liu: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
shfaithy: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Anki98765: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
MassBassLol: RT @roydanroy: Any optimization experts out there willing to weigh in? https://t.co/tG1JsqGLOg https://t.co/Kx4lQ2bIaL
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 5
Total Words: 14525
Unqiue Words: 2449

2.163 Mikeys
#2. Quantum-inspired low-rank stochastic regression with logarithmic dependence on the dimension
András Gilyén, Seth Lloyd, Ewin Tang
We construct an efficient classical analogue of the quantum matrix inversion algorithm (HHL) for low-rank matrices. Inspired by recent work of Tang, assuming length-square sampling access to input data, we implement the pseudoinverse of a low-rank matrix and sample from the solution to the problem $Ax=b$ using fast sampling techniques. We implement the pseudo-inverse by finding an approximate singular value decomposition of $A$ via subsampling, then inverting the singular values. In principle, the approach can also be used to apply any desired "smooth" function to the singular values. Since many quantum algorithms can be expressed as a singular value transformation problem, our result suggests that more low-rank quantum algorithms can be effectively "dequantised" into classical length-square sampling algorithms.
more | pdf | html
Figures
None.
Tweets
lukOlejnik: Interesting result. A specific algorithm for a quantum computer ("low-rank stochastic regression") found to be efficiently solved by non-quantum computers. Not yet understood well which problems are best solved with quantum computers. https://t.co/ugGX81ga6m https://t.co/pr8CAPK6uy
QuantumMemeing: Fig. 1.26: A meme [1]. [1] Quantum Computing Memes for QMA-Complete Teens, Studies in Ancient Greek and Syriac Memeing (2018). Ewin Tang destroying the hopes and dreams of QMLers everywhere. Check out the paper here: https://t.co/jNo9XaPsRI https://t.co/oQwzeVkLyS
yuyu_hf: また量子情報◯すマンの新作だ、、、 https://t.co/Ebas1imqAV
fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
ComputerPapers: Quantum-inspired low-rank stochastic regression with logarithmic dependence on the dimension. https://t.co/jDZCT9f2us
ewintang: New twin papers on quantum-inspired algorithms for low-rank matrix inversion: https://t.co/3T6Xx9G3hy and https://t.co/tUPCxbq0UA
octonion: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
makoto0218ne56: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
spidermanzano: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
Lukasaoz: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
nick_farina: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
postquantum: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
katelovesneuro: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
RichFelker: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
durumcrustulum: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
rdviii: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
gejikeiji: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
rbtcollins: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
iKodack: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
kamakiri_ys: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
jpdowling: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
MGimenoSegovia: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
aquintex: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
henryquantum: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
BulentKIzILtan: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
amy8492: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
matt_reagor: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
rayohauno: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
cocori_aqua: RT @yuyu_hf: また量子情報◯すマンの新作だ、、、 https://t.co/Ebas1imqAV
cocori_aqua: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
ilyaraz2: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
TilmaLabs: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
yoshi_and_aki: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
yoshi_and_aki: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
quantumbtc: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
AlyTarekIbrahim: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
ywyamashiro: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
wjmzbmr1: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
world_fantasia: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
akinori_kawachi: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
gshartnett: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
Deepneuron: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
DhammaKimpara: RT @QuantumHazzard: Wow -- the famous HHL algorithm doesn't actually need a quantum computer! https://t.co/5ZduTl0vcZ Another quantum algor…
ons_yy: RT @fgksk: うぉ。第三弾か。https://t.co/oMrsU9OB02
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 5438
Unqiue Words: 1428

2.066 Mikeys
#3. Embracing the Laws of Physics: Three Reversible Models of Computation
Jacques Carette, Roshan P. James, Amr Sabry
Our main models of computation (the Turing Machine and the RAM) make fundamental assumptions about which primitive operations are realizable. The consensus is that these include logical operations like conjunction, disjunction and negation, as well as reading and writing to memory locations. This perspective conforms to a macro-level view of physics and indeed these operations are realizable using macro-level devices involving thousands of electrons. This point of view is however incompatible with quantum mechanics, or even elementary thermodynamics, as both imply that information is a conserved quantity of physical processes, and hence of primitive computational operations. Our aim is to re-develop foundational computational models that embraces the principle of conservation of information. We first define what conservation of information means in a computational setting. We emphasize that computations must be reversible transformations on data. One can think of data as modeled using topological spaces and programs as modeled...
more | pdf | html
Figures
Tweets
HNTweets: Embracing the Laws of Physics: Three Reversible Models of Computation: https://t.co/92Law8R96K Comments: https://t.co/aYLFpwEAS1
arxiv_org: Embracing the Laws of Physics: Three Reversible Models of Computation. https://t.co/VzRixLj4Bq https://t.co/RQqHnjM624
Aldana_Angel: https://t.co/hPIJbhsNIb Embracing the Laws of Physics: Three Reversible Models of Computation Jacques Carette, Roshan P. James, Amr Sabry
sigfpe: One for the to-read list: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/KgRMHc4fri
StephenPiment: Embracing the Laws of Physics: Three Reversible Models of Computation (using Curry-Howard) https://t.co/Vh8Tz6W424
jmsunico: sn-news: #sw #dev #maths Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/ElxzDRwG2C
hn_frontpage: Embracing the Laws of Physics: Three Reversible Models of Computation L: https://t.co/OJVUK7CPhd C: https://t.co/z6Hu4842H4
kov4l3nko: Hmmm... just in time:)[1811.03678] Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/OM1bq2b06X
andrewfnewman: "Programs as Reversible Deformations" https://t.co/AHTE3QGsBE something to get into if you're not dealing with JavaScript.
kushnerbomb: gud paper, much more type theory than I expected https://t.co/VxCs4zEmY0
angsuman: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/mMLEyIFsUZ
betterhn50: 51 – Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/JsPmBCjHtD
QuantumPapers: Embracing the Laws of Physics: Three Reversible Models of Computation. https://t.co/NNq442eSaD
joshtronic: Embracing the Laws of Physics: Three Reversible Models of Computation - https://t.co/IF5rLlIQdu
dJdU: “Embracing the Laws of Physics: Three Reversible Models of Computation” https://t.co/AkPI6aRpga
hackernewsrobot: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/Atno35U9Mi
jackhidary: RT @StephenPiment: Embracing the Laws of Physics: Three Reversible Models of Computation (using Curry-Howard) https://t.co/Vh8Tz6W424
Juan_A_Lleo: RT @StephenPiment: Embracing the Laws of Physics: Three Reversible Models of Computation (using Curry-Howard) https://t.co/Vh8Tz6W424
paul_snively: RT @sigfpe: One for the to-read list: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/KgRMHc4fri
PLT_cheater: RT @sigfpe: One for the to-read list: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/KgRMHc4fri
Priceeqn: RT @StephenPiment: Embracing the Laws of Physics: Three Reversible Models of Computation (using Curry-Howard) https://t.co/Vh8Tz6W424
jjcarett2: RT @arxiv_cslo: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/nGVvjPyTqm
maxsnew: RT @arxiv_cslo: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/nGVvjPyTqm
pauldhoward: RT @StephenPiment: Embracing the Laws of Physics: Three Reversible Models of Computation (using Curry-Howard) https://t.co/Vh8Tz6W424
SandMouth: RT @arxiv_cslo: Embracing the Laws of Physics: Three Reversible Models of Computation https://t.co/nGVvjPyTqm
shubh_300595: RT @arxiv_org: Embracing the Laws of Physics: Three Reversible Models of Computation. https://t.co/VzRixLj4Bq https://t.co/RQqHnjM624
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 16963
Unqiue Words: 4120

2.047 Mikeys
#4. Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction
Yannick Suter, Alain Jungo, Michael Rebsamen, Urspeter Knecht, Evelyn Herrmann, Roland Wiest, Mauricio Reyes
Deep learning for regression tasks on medical imaging data has shown promising results. However, compared to other approaches, their power is strongly linked to the dataset size. In this study, we evaluate 3D-convolutional neural networks (CNNs) and classical regression methods with hand-crafted features for survival time regression of patients with high grade brain tumors. The tested CNNs for regression showed promising but unstable results. The best performing deep learning approach reached an accuracy of 51.5% on held-out samples of the training set. All tested deep learning experiments were outperformed by a Support Vector Classifier (SVC) using 30 radiomic features. The investigated features included intensity, shape, location and deep features. The submitted method to the BraTS 2018 survival prediction challenge is an ensemble of SVCs, which reached a cross-validated accuracy of 72.2% on the BraTS 2018 training set, 57.1% on the validation set, and 42.9% on the testing set. The results suggest that more training data...
more | pdf | html
Figures
Tweets
BrundageBot: Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction. Yannick Suter, Alain Jungo, Michael Rebsamen, Urspeter Knecht, Evelyn Herrmann, Roland Wiest, and Mauricio Reyes https://t.co/VEQmuPViqQ
arxivml: "Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction", Yannick Suter, Alain Jungo… https://t.co/wzHvywfqp5
nmfeeds: [CV] https://t.co/VNeKC4aP2L Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction. Deep l...
nmfeeds: [O] https://t.co/VNeKC4aP2L Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction. Deep le...
Memoirs: Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction. https://t.co/MwW4XvorCy
arxiv_cscv: Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction https://t.co/vUjG6qv8RM
arxiv_cscv: Deep Learning versus Classical Regression for Brain Tumor Patient Survival Prediction https://t.co/vUjG6qv8RM
Github
None.
Youtube
None.
Other stats
Sample Sizes : [1353, 163]
Authors: 7
Total Words: 5149
Unqiue Words: 1867

2.028 Mikeys
#5. Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?
Hongming Shan, Atul Padole, Fatemeh Homayounieh, Uwe Kruger, Ruhani Doda Khera, Chayanin Nitiwarangkul, Mannudeep K. Kalra, Ge Wang
Commercial iterative reconstruction techniques on modern CT scanners target radiation dose reduction but there are lingering concerns over their impact on image appearance and low contrast detectability. Recently, machine learning, especially deep learning, has been actively investigated for CT. Here we design a novel neural network architecture for low-dose CT (LDCT) and compare it with commercial iterative reconstruction methods used for standard of care CT. While popular neural networks are trained for end-to-end mapping, driven by big data, our novel neural network is intended for end-to-process mapping so that intermediate image targets are obtained with the associated search gradients along which the final image targets are gradually reached. This learned dynamic process allows to include radiologists in the training loop to optimize the LDCT denoising workflow in a task-specific fashion with the denoising depth as a key parameter. Our progressive denoising network was trained with the Mayo LDCT Challenge Dataset, and tested...
more | pdf | html
Figures
Tweets
arxiv_org: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/sweZ5cYgEi https://t.co/1P4xBYl6AV
BrundageBot: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. Hongming Shan, Atul Padole, Fatemeh Homayounieh, Uwe Kruger, Ruhani Doda Khera, Chayanin Nitiwarangkul, Mannudeep K. Kalra, and Ge Wang https://t.co/zJFXBguFQ9
arxivml: "Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?", Hongming Shan, Atul Padole, Fate… https://t.co/0JOrx9w9em
nmfeeds: [O] https://t.co/4qMqzjGXo3 Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. Commercial it...
nmfeeds: [CV] https://t.co/4qMqzjGXo3 Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. Commercial i...
kurt_koo: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/AcM8ibYFhp #machinelearning #DeepLearning #medicalimageanalysis https://t.co/ZRxncX3lD7
arxiv_cscv: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/gJSgzLkST4
PhysicsPaper: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/qDqSlocz6q
ComputerPapers: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/m6JQnfIfkR
machinelearn_d: RT @kurt_koo: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/AcM8ibYFhp #machinelearning #D…
TechNowOrNever: RT @kurt_koo: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/AcM8ibYFhp #machinelearning #D…
msarozz: RT @kurt_koo: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/AcM8ibYFhp #machinelearning #D…
Zahid_Akhtar: RT @arxiv_org: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/sweZ5cYgEi https://t.co/1P4xBY…
theChrisChua: RT @kurt_koo: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods? https://t.co/AcM8ibYFhp #machinelearning #D…
shubh_300595: RT @arxiv_org: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/sweZ5cYgEi https://t.co/1P4xBY…
Yubram11: RT @arxiv_org: Can Deep Learning Outperform Modern Commercial CT Image Reconstruction Methods?. https://t.co/sweZ5cYgEi https://t.co/1P4xBY…
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 8
Total Words: 8205
Unqiue Words: 2491

2.027 Mikeys
#6. TED: Teaching AI to Explain its Decisions
Noel C. F. Codella, Michael Hind, Karthikeyan Natesan Ramamurthy, Murray Campbell, Amit Dhurandhar, Kush R. Varshney, Dennis Wei, Aleksandra Mojsilovic
Artificial intelligence systems are being increasingly deployed due to their potential to increase the efficiency, scale, consistency, fairness, and accuracy of decisions. However, as many of these systems are opaque in their operation, there is a growing demand for such systems to provide explanations for their decisions. Conventional approaches to this problem attempt to expose or discover the inner workings of a machine learning model with the hope that the resulting explanations will be meaningful to the consumer. In contrast, this paper suggests a new approach to this problem. It introduces a simple, practical framework, called Teaching Explanations for Decisions (TED), that provides meaningful explanations that match the mental model of the consumer. We illustrate the generality and effectiveness of this approach with two different examples, resulting in highly accurate explanations with no loss of prediction accuracy for these two examples.
more | pdf | html
Figures
Tweets
BrundageBot: TED: Teaching AI to Explain its Decisions. Noel C. F. Codella, Michael Hind, Karthikeyan Natesan Ramamurthy, Murray Campbell, Amit Dhurandhar, Kush R. Varshney, Dennis Wei, and Aleksandra Mojsilovic https://t.co/Dju2Fzo7ZW
arxivml: "TED: Teaching AI to Explain its Decisions", Noel C. F. Codella, Michael Hind, Karthikeyan Natesan Ramamurthy, Murr… https://t.co/kmuAkPoShn
nmfeeds: [AI] https://t.co/hGH2tkmKCp TED: Teaching AI to Explain its Decisions. Artificial intelligence systems are being increasi...
nmfeeds: [O] https://t.co/hGH2tkmKCp TED: Teaching AI to Explain its Decisions. Artificial intelligence systems are being increasin...
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 8
Total Words: 6175
Unqiue Words: 2074

2.025 Mikeys
#7. Langevin-gradient parallel tempering for Bayesian neural learning
Rohitash Chandra, Konark Jain, Ratneel V. Deo, Sally Cripps
Bayesian neural learning feature a rigorous approach to estimation and uncertainty quantification via the posterior distribution of weights that represent knowledge of the neural network. This not only provides point estimates of optimal set of weights but also the ability to quantify uncertainty in decision making using the posterior distribution. Markov chain Monte Carlo (MCMC) techniques are typically used to obtain sample-based estimates of the posterior distribution. However, these techniques face challenges in convergence and scalability, particularly in settings with large datasets and network architectures. This paper address these challenges in two ways. First, parallel tempering is used used to explore multiple modes of the posterior distribution and implemented in multi-core computing architecture. Second, we make within-chain sampling schemes more efficient by using Langevin gradient information in forming Metropolis-Hastings proposal distributions. We demonstrate the techniques using time series prediction and...
more | pdf | html
Figures
Tweets
arxivml: "Langevin-gradient parallel tempering for Bayesian neural learning", Rohitash Chandra, Konark Jain, Ratneel V. Deo,… https://t.co/MahsHSaACO
nmfeeds: [AI] https://t.co/2GPetdoCV1 Langevin-gradient parallel tempering for Bayesian neural learning. Bayesian neural learning f...
nmfeeds: [O] https://t.co/2GPetdoCV1 Langevin-gradient parallel tempering for Bayesian neural learning. Bayesian neural learning fe...
SciFi: Langevin-gradient parallel tempering for Bayesian neural learning. https://t.co/SaK4v7dqHP
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 4
Total Words: 9717
Unqiue Words: 3079

2.025 Mikeys
#8. Focusing on the Big Picture: Insights into a Systems Approach to Deep Learning for Satellite Imagery
Ritwik Gupta, Carson D. Sestili, Javier A. Vazquez-Trejo, Matthew E. Gaston
Deep learning tasks are often complicated and require a variety of components working together efficiently to perform well. Due to the often large scale of these tasks, there is a necessity to iterate quickly in order to attempt a variety of methods and to find and fix bugs. While participating in IARPA's Functional Map of the World challenge, we identified challenges along the entire deep learning pipeline and found various solutions to these challenges. In this paper, we present the performance, engineering, and deep learning considerations with processing and modeling data, as well as underlying infrastructure considerations that support large-scale deep learning tasks. We also discuss insights and observations with regard to satellite imagery and deep learning for image classification.
more | pdf | html
Figures
Tweets
arxivml: "Focusing on the Big Picture: Insights into a Systems Approach to Deep Learning for Satellite Imagery", Ritwik Gupt… https://t.co/0v7mlYLfja
nmfeeds: [CV] https://t.co/xOERMe0vbU Focusing on the Big Picture: Insights into a Systems Approach to Deep Learning for Satellite ...
nmfeeds: [O] https://t.co/xOERMe0vbU Focusing on the Big Picture: Insights into a Systems Approach to Deep Learning for Satellite I...
arxiv_cscv: Focusing on the Big Picture: Insights into a Systems Approach to Deep Learning for Satellite Imagery https://t.co/trmkACLWGk
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 4
Total Words: 4282
Unqiue Words: 1626

2.024 Mikeys
#9. Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers
Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang
Neural networks have great success in many machine learning applications, but the fundamental learning theory behind them remains largely unsolved. Learning neural networks is NP-hard, but in practice, simple algorithms like stochastic gradient descent (SGD) often produce good solutions. Moreover, it is observed that overparameterization --- designing networks whose number of parameters is larger than statistically needed to perfectly fit the data --- improves both optimization and generalization, appearing to contradict traditional learning theory. In this work, we extend the theoretical understanding of two and three-layer neural networks in the overparameterized regime. We prove that, using overparameterized neural networks, one can (improperly) learn some notable hypothesis classes, including two and three-layer neural networks with fewer parameters. Moreover, the learning process can be simply done by SGD or its variants in polynomial time using polynomially many samples. We also show that for a fixed sample size, the...
more | pdf | html
Figures
None.
Tweets
arxivml: "Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers", Zeyuan Allen-Zhu, Yuan… https://t.co/eQUwx5h3Ba
nmfeeds: [NE] https://t.co/P2Na7yZNzn Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers. Ne...
nmfeeds: [O] https://t.co/P2Na7yZNzn Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers. Neu...
Soul: Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers. https://t.co/X0WcsIHsyM
Github
None.
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 26203
Unqiue Words: 4261

2.024 Mikeys
#10. Learning Energy Based Inpainting for Optical Flow
Christoph Vogel, Patrick Knöbelreiter, Thomas Pock
Modern optical flow methods are often composed of a cascade of many independent steps or formulated as a black box neural network that is hard to interpret and analyze. In this work we seek for a plain, interpretable, but learnable solution. We propose a novel inpainting based algorithm that approaches the problem in three steps: feature selection and matching, selection of supporting points and energy based inpainting. To facilitate the inference we propose an optimization layer that allows to backpropagate through 10K iterations of a first-order method without any numerical or memory problems. Compared to recent state-of-the-art networks, our modular CNN is very lightweight and competitive with other, more involved, inpainting based methods.
more | pdf | html
Figures
Tweets
arxiv_org: Learning Energy Based Inpainting for Optical Flow. https://t.co/pF2kJFzBDy https://t.co/KaiJkbTMWK
arxivml: "Learning Energy Based Inpainting for Optical Flow", Christoph Vogel, Patrick Knöbelreiter, Thomas Pock https://t.co/FKWyAipr6i
nmfeeds: [CV] https://t.co/YCOJyPDvU4 Learning Energy Based Inpainting for Optical Flow. Modern optical flow methods are often comp...
nmfeeds: [O] https://t.co/YCOJyPDvU4 Learning Energy Based Inpainting for Optical Flow. Modern optical flow methods are often compo...
arxiv_cscv: Learning Energy Based Inpainting for Optical Flow https://t.co/YxHtE4GJ6j
arxiv_cscv: Learning Energy Based Inpainting for Optical Flow https://t.co/YxHtE4YjXR
ComputerPapers: Learning Energy Based Inpainting for Optical Flow. https://t.co/WtlCCOJZ1p
ugredium: RT @arxiv_org: Learning Energy Based Inpainting for Optical Flow. https://t.co/pF2kJFzBDy https://t.co/KaiJkbTMWK
shubh_300595: RT @arxiv_org: Learning Energy Based Inpainting for Optical Flow. https://t.co/pF2kJFzBDy https://t.co/KaiJkbTMWK
Github

The repository holds several custom network layers. Some of which were used in my recent optical flow project: Learning Energy Based Inpainting for Optical Flow.

Repository: CustomNetworkLayers
User: vogechri
Language: Cuda
Stargazers: 4
Subscribers: 3
Forks: 0
Open Issues: 0
Youtube
None.
Other stats
Sample Sizes : None.
Authors: 3
Total Words: 7244
Unqiue Words: 2353

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 56,475 papers.

Search
Sort results based on if they are interesting or reproducible.
Interesting
Reproducible
Categories
All
Astrophysics
Cosmology and Nongalactic Astrophysics
Earth and Planetary Astrophysics
Astrophysics of Galaxies
High Energy Astrophysical Phenomena
Instrumentation and Methods for Astrophysics
Solar and Stellar Astrophysics
Condensed Matter
Disordered Systems and Neural Networks
Mesoscale and Nanoscale Physics
Materials Science
Other Condensed Matter
Quantum Gases
Soft Condensed Matter
Statistical Mechanics
Strongly Correlated Electrons
Superconductivity
Computer Science
Artificial Intelligence
Hardware Architecture
Computational Complexity
Computational Engineering, Finance, and Science
Computational Geometry
Computation and Language
Cryptography and Security
Computer Vision and Pattern Recognition
Computers and Society
Databases
Distributed, Parallel, and Cluster Computing
Digital Libraries
Discrete Mathematics
Data Structures and Algorithms
Emerging Technologies
Formal Languages and Automata Theory
General Literature
Graphics
Computer Science and Game Theory
Human-Computer Interaction
Information Retrieval
Information Theory
Machine Learning
Logic in Computer Science
Multiagent Systems
Multimedia
Mathematical Software
Numerical Analysis
Neural and Evolutionary Computing
Networking and Internet Architecture
Other Computer Science
Operating Systems
Performance
Programming Languages
Robotics
Symbolic Computation
Sound
Software Engineering
Social and Information Networks
Systems and Control
Economics
Econometrics
General Economics
Theoretical Economics
Electrical Engineering and Systems Science
Audio and Speech Processing
Image and Video Processing
Signal Processing
General Relativity and Quantum Cosmology
General Relativity and Quantum Cosmology
High Energy Physics - Experiment
High Energy Physics - Experiment
High Energy Physics - Lattice
High Energy Physics - Lattice
High Energy Physics - Phenomenology
High Energy Physics - Phenomenology
High Energy Physics - Theory
High Energy Physics - Theory
Mathematics
Commutative Algebra
Algebraic Geometry
Analysis of PDEs
Algebraic Topology
Classical Analysis and ODEs
Combinatorics
Category Theory
Complex Variables
Differential Geometry
Dynamical Systems
Functional Analysis
General Mathematics
General Topology
Group Theory
Geometric Topology
History and Overview
Information Theory
K-Theory and Homology
Logic
Metric Geometry
Mathematical Physics
Numerical Analysis
Number Theory
Operator Algebras
Optimization and Control
Probability
Quantum Algebra
Rings and Algebras
Representation Theory
Symplectic Geometry
Spectral Theory
Statistics Theory
Mathematical Physics
Mathematical Physics
Nonlinear Sciences
Adaptation and Self-Organizing Systems
Chaotic Dynamics
Cellular Automata and Lattice Gases
Pattern Formation and Solitons
Exactly Solvable and Integrable Systems
Nuclear Experiment
Nuclear Experiment
Nuclear Theory
Nuclear Theory
Physics
Accelerator Physics
Atmospheric and Oceanic Physics
Applied Physics
Atomic and Molecular Clusters
Atomic Physics
Biological Physics
Chemical Physics
Classical Physics
Computational Physics
Data Analysis, Statistics and Probability
Physics Education
Fluid Dynamics
General Physics
Geophysics
History and Philosophy of Physics
Instrumentation and Detectors
Medical Physics
Optics
Plasma Physics
Popular Physics
Physics and Society
Space Physics
Quantitative Biology
Biomolecules
Cell Behavior
Genomics
Molecular Networks
Neurons and Cognition
Other Quantitative Biology
Populations and Evolution
Quantitative Methods
Subcellular Processes
Tissues and Organs
Quantitative Finance
Computational Finance
Economics
General Finance
Mathematical Finance
Portfolio Management
Pricing of Securities
Risk Management
Statistical Finance
Trading and Market Microstructure
Quantum Physics
Quantum Physics
Statistics
Applications
Computation
Methodology
Machine Learning
Other Statistics
Statistics Theory
Feedback
Online
Stats
Tracking 56,475 papers.