chunk_uid
stringlengths
40
40
chunk_type
stringclasses
2 values
chunk_index
int64
0
6.71k
total_chunks
int64
1
6.71k
section_title
stringlengths
1
157
embed_text
stringlengths
1
83.3k
spans
dict
paper_doi
stringlengths
0
63
paper_id_arxiv
stringlengths
9
16
title
stringlengths
7
245
authors
listlengths
1
768
categories
listlengths
1
7
year
int64
2k
2.02k
language
stringclasses
2 values
discipline
stringclasses
8 values
dense_vector
listlengths
1.02k
1.02k
6c70b0b3fd558a29c96d36641f722e930c986342
subsection
26
39
Adaptive Algorithm for Theorem
Moreover, \mathcal {L} = 0 \!\Rightarrow \! \widetilde{r}_m = 1 and so the reference indicators introduced here for the general case devolve into those of the previous subsection for f independent of u. Therefore, we conclude that the two adaptive algorithms are the same for this type of data. \end{equation}\begin{} \b...
{ "cite_spans": [] }
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.018078478053212166, 0.03325829654932022, -0.06413663923740387, 0.004397570621222258, 0.0071017141453921795, -0.045249588787555695, 0.009596086107194424, 0.0038693281821906567, 0.035516198724508286, 0.006075741723179817, 0.012448976747691631, 0.009908836334943771, -0.029352735728025436, ...
38b5b9530b25f4f15916bcd3c5befca713693fe3
subsection
27
39
Numerical Experiments
We consider an implementation of the adaptive algorithm of the previous section through an application of the deal.II finite element library , . In order to facilitate a comparison between the L^\infty L^\infty estimator of Theorem REF and the L^2H^1 estimator of , we consider Example 1 and Example 3 of but under the a...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 144, "openalex_id": "", "raw": "Wolfgang Bangerth, Ralf Hartmann, and Guido Kanschat, deal.ii – a general-purpose object-oriented finite element library, ACM Transactions on Mathematical Software (TOMS) 33 (2007), no. 4, 24.", ...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.03704969212412834, 0.023499391973018646, -0.0232399832457304, -0.005226325709372759, -0.0059091816656291485, -0.0005541049758903682, 0.003170129144564271, -0.002920257393270731, 0.03454715758562088, 0.03059498779475689, -0.045472849160432816, -0.014252228662371635, -0.04358069226145744, ...
e635233936efd09d6a987beee3dcb37540280b62
subsection
28
39
Example 1
Let \Omega = (-8,8)^2, a = 1, f(u) = u^2 and choose the initial condition to be the Gaussian blob given by u_0(x,y) = 10\exp (-2x^2-2y^2). The blow-up set for this example consists of only a single point (the origin) making it spatially uncomplicated which allows us to focus solely on the temporal asymptotics. Now, sin...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 1589, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM ...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.07423373311758041, 0.023060685023665428, -0.03794100880622864, -0.011255629360675812, -0.012476579286158085, -0.027685033157467842, -0.000015679186617489904, 0.03198888152837753, 0.036903202533721924, 0.03583487123250961, -0.021412404254078865, 0.03117999993264675, -0.0010034679435193539,...
8cdd51c00b2464eff24504ca2329ee3daa16c595
subsection
29
39
Example 1
Using this approximation to T_{\infty }, we take the data from Table REF and plot the distance from the blow-up time |T-T_{\infty }| versus the total number of time steps N in Figure REF . The plot shows that for this example we have\begin{aligned}|T_{\infty } - T({\tt ttol^+}, N)| \propto N^{-3 \slash 4}. \end{aligned...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 390, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM J...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.048074256628751755, 0.013460791669785976, -0.025090184062719345, 0.009134109131991863, 0.009820884093642235, -0.054209448397159576, 0.0012371490010991693, 0.020511683076620102, 0.0552167184650898, 0.03195793554186821, -0.007611757609993219, 0.033300962299108505, -0.01020242553204298, 0....
c75e8e94eebad05201f0df9e44bcfc8d56f63533
subsection
30
39
Example 1
\end{split}Therefore, we have \mathcal {L}(|v_1|,|v_2|) = |v_1| + |v_2| in (REF ) and so \delta _m (if it exists) is the smallest root of the function \varphi _m:[1,\infty ) \rightarrow \mathbb {R} given by\begin{split} \varphi _m(\delta ) = 1+ \delta \!\left[\int _{I_m} L(s, \delta ) \,\text{d}s -1 \right]\! = 1 + \d...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 1131, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM ...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.07366587966680527, 0.029325976967811584, -0.011008684523403645, -0.0027617074083536863, -0.021925821900367737, 0.0024451029021292925, 0.011268071830272675, 0.031858813017606735, 0.04711686447262764, 0.03906061500310898, -0.005656922701746225, 0.01203860342502594, -0.005374649073928595, ...
97b4c38e6e5ae7e8fc5ddaf7db48951bb999ff81
subsection
31
39
Example 1
From the results, given in Figure REF , we deduce the asymptotic estimate\begin{aligned}||U(t)|| \propto |t - T_{\infty }|^{-1}, \end{aligned}which is consistent with the asymptotics of the exact solution , suggesting that our numerical solution is reasonable. Moreover, we expect (cf. Corollary 4.3 of ) that in the spa...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1002/(sici)1097-0312(199802)51:2<139::aid-cpa2>3.0.co;2-c", "end": 261, "openalex_id": "https://openalex.org/W2026854825", "raw": "Frank Merle and Hatem Zaag, Optimal estimates for blowup rate and behavior for nonlinear heat equations, C...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.033170390874147415, 0.002549954690039158, -0.03228543698787689, -0.007266512140631676, 0.0126410610973835, -0.047238051891326904, 0.018004167824983597, 0.018568703904747963, 0.05968838930130005, 0.022978199645876884, 0.020781081169843674, 0.021940670907497406, -0.007647956721484661, 0.0...
644c336fb5f7c04f1add1782ef59cddcfa22a80b
subsection
32
39
Example 2
Let \Omega = (-8,8)^2, a = 1, f(u) = u^2 and the “volcano” type initial condition be given by u_0(x,y) = 10(x^2+y^2)\exp (-0.5x^2-0.5y^2). The blow-up set for this example is a circle centered on the origin making this example a good test of the spatial capabilities of the adaptive algorithm as many degrees of freedom ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 1393, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM ...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.048786941915750504, 0.020320264622569084, -0.04814621061086655, -0.02291369065642357, 0.004519428126513958, -0.002328363712877035, 0.009069367311894894, 0.022150918841362, 0.0465596467256546, 0.01762004941701889, -0.02172376587986946, 0.009115133434534073, -0.022959457710385323, 0.02742...
48e58d4fd31af7b68b6b3e518db9fed79991fb65
subsection
33
39
Example 2
For visualization purposes, we also display profile views of the finite element solution at times t=0 and t=T in Figure REF .Let \Omega = (-8,8)^2, a = 1, f(u) = u^2 and the “volcano” type initial condition be given by u_0(x,y) = 10(x^2+y^2)\exp (-0.5x^2-0.5y^2). The blow-up set for this example is a circle centered on...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 1518, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM ...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.03533045947551727, 0.014103199355304241, -0.03667289391160011, -0.010373365134000778, 0.004046374931931496, 0.008969909511506557, 0.026330040767788887, 0.01818389818072319, 0.03783227130770683, 0.03472026064991951, -0.010205560363829136, -0.006815148051828146, -0.007783837150782347, 0.0...
0c19f6bad025b678d44b1dcc6b0447bd7c6bf4a0
subsection
34
39
Example 3
In this example, we consider a nonlinear parabolic problem from , . We set \Omega = (0,1)^2, T = 0.75, a = 0.001, f(t,u) = \sin (t) - u^4 and u_0(x,y)=xy(x-1)(y-1). The solution is initially unremarkable but as time evolves it begins to exhibit boundary layers through the influence of the diffusion and the forcing term...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1093/imanum/drw049", "end": 67, "openalex_id": "https://openalex.org/W2525999793", "raw": "Mario Amrein and Thomas P. Wihler, An adaptive space-time Newton–Galerkin approach for semilinear singularly perturbed parabolic evolution equatio...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.046963389962911606, 0.005542381666600704, -0.026716340333223343, -0.04140956327319145, 0.025236336514353752, 0.024915924295783043, 0.0161274541169405, 0.019835090264678, 0.07598365098237991, 0.0431184321641922, -0.0016001577023416758, -0.005656815133988857, -0.01327425241470337, 0.01176...
dc14d65fb5fbc55fd7361eda28f9180a2dc246f3
subsection
35
39
Example 3
In order to quantify the rate of convergence of the estimator in space, we first choose a small temporal refinement tolerance {\tt ttol}^+ so that the size of the temporal contribution to the estimator is negligible; we then gradually decrease the spatial refinement tolerance {\tt stol}^+ for polynomials of degree thre...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1093/imanum/drw049", "end": 1142, "openalex_id": "https://openalex.org/W2525999793", "raw": "Mario Amrein and Thomas P. Wihler, An adaptive space-time Newton–Galerkin approach for semilinear singularly perturbed parabolic evolution equat...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.05213933065533638, 0.009526334702968597, -0.028586631640791893, -0.028434088453650475, 0.004332232289016247, 0.03691549971699715, 0.014316197484731674, 0.00769962789490819, 0.06290889531373978, 0.03825788199901581, -0.006429703906178474, 0.010472103953361511, -0.0030642154160887003, 0.0...
d609021445dfe89d2c8e544a529b4d75ddd58977
subsection
36
39
Example 3
To do this, we choose a large polynomial degree and a small spatial refinement tolerance {\tt stol}^+ so that the spatial contribution to the estimator is negligible; we then gradually reduce the temporal refinement tolerance {\tt ttol}^+ in order to observe the rate of convergence of the estimator in time. Next, we pl...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 910, "openalex_id": "", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, and Stephen Metcalfe, Adaptive discontinuous Galerkin methods for nonstationary convection–diffusion problems, IMA Journal of Numerical Analysis (2013), drt0...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.036216069012880325, 0.0019517241744324565, -0.04805417358875275, -0.025964509695768356, 0.01020579133182764, 0.004702444188296795, -0.01041173841804266, -0.01893181912600994, 0.06706227362155914, 0.04558281600475311, -0.01827584207057953, 0.018047012388706207, -0.00883281510323286, 0.00...
30c761e64f6852a61ba25d926871f01844ddd532
subsection
37
39
Conclusions
We derived a conditional L^{\infty }L^{\infty } a posteriori error bound (Theorem REF ) for the IMEX discretization (REF ) of the semilinear heat equation (REF ) with general local Lipschitz nonlinearity (REF ). Our numerical experiments indicate that the proposed estimator outperforms the L^2 H^1 estimator of with res...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/16m106073x", "end": 360, "openalex_id": "https://openalex.org/W1524366444", "raw": "Andrea Cangiani, Emmanuil H. Georgoulis, Irene Kyza, and Stephen Metcalfe, Adaptivity and blow-up detection for nonlinear evolution problems, SIAM J...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.05445661023259163, 0.01623930223286152, -0.022954802960157394, -0.02437421679496765, -0.0069444384425878525, -0.010660857893526554, 0.005017547868192196, 0.02428264170885086, 0.024725254625082016, 0.01862025260925293, -0.052411433309316635, -0.006784182507544756, -0.03461535647511482, 0...
f4ef0ea181d152a83b91b88a21954b63cf8fc79a
subsection
38
39
Conclusions
The slow convergence in Example 1 can be explained by the initial condition not having a “compatible profile” with the blow-up, that is, this choice of initial condition causes the solution to be significantly influenced by the laplacian early on (this can be seen in Figure REF by noting that the numerical solution onl...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1137/070708792", "end": 744, "openalex_id": "https://openalex.org/W1973664641", "raw": "Alan Demlow, Omar Lakkis, and Charalambos Makridakis, A posteriori error estimates in the maximum norm for parabolic problems, SIAM Journal on Numeri...
1802.07757
Pointwise a posteriori error bounds for blow-up in the semilinear heat equation
[ "Irene Kyza", "Stephen Metcalfe" ]
[ "math.NA" ]
2,018
en
Mathematics
[ -0.023051386699080467, 0.027109406888484955, -0.02619406394660473, -0.040671736001968384, -0.01064848992973566, -0.0034668613225221634, -0.002595378551632166, 0.04079378396272659, 0.03826133534312248, 0.034935589879751205, -0.021251210942864418, -0.013562330976128578, -0.010640861466526985, ...
07857de7b4724aa75619ad4be826754cd8f62d9a
abstract
0
30
Abstract
We study stochastic algorithms for solving nonconvex optimization problems with a convex yet possibly nonsmooth regularizer, which find wide applications in many practical machine learning applications. However, compared to asynchronous parallel stochastic gradient descent (AsynSGD), an algorithm targeting smooth optim...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03524276986718178, 0.013425816781818867, -0.025387000292539597, -0.02898755855858326, -0.010565202683210373, 0.017346765846014023, 0.030742069706320763, -0.0031752819195389748, 0.03203888237476349, 0.06725113838911057, -0.02746189758181572, 0.026851633563637733, 0.0009263622923754156, -...
5637b9a4d690e173d269a49a3a9a6ff34bcf7e69
subsection
1
30
Introduction
With rapidly growing data volumes and variety, the need to scale up machine learning has sparked broad interests in developing efficient parallel optimization algorithms. A typical parallel optimization algorithm usually decomposes the original problem into multiple subproblems, each handled by a worker node. Each work...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 674, "openalex_id": "", "raw": "F. Niu, B. Recht, C. Re, and S. Wright. Hogwild: A lock-free approach to parallelizing stochastic gradient descent. In Advances in Neural Information Processing Systems, pages 693–701, 2011.", ...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.0290891844779253, 0.003359526162967086, -0.03488870710134506, -0.03546866029500961, -0.020222019404172897, 0.04062718152999878, 0.015292425639927387, 0.01555187813937664, 0.018466901034116745, 0.03772742301225662, -0.046609848737716675, 0.009836295619606972, -0.014567485079169273, 0.030...
69160c56716745f9e3f5275d2619312657b7b373
subsection
2
30
Introduction
An asynchronous parallel proximal gradient method has been presented in and has been shown to converge to stationary points for nonconvex problems. However, has essentially proposed a non-stochastic algorithm and has not provided its convergence rate.In this paper, we propose and analyze an asynchronous parallel proxim...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 148, "openalex_id": "https://openalex.org/W2127941149", "raw": "M. Li, D. G. Andersen, A. J. Smola, and K. Yu. Communication efficient distributed machine learning with the parameter server. In Advances in Neural Information Proce...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.058228399604558945, -0.014167995192110538, -0.03463796153664589, -0.04602118954062462, 0.0014505600556731224, 0.012764166109263897, 0.03114364854991436, 0.011650257743895054, 0.030426474288105965, 0.036957334727048874, -0.03952084854245186, 0.01649499498307705, -0.00900281872600317, 0.0...
9e9d7f583fc4f98bd7c1d317d8a6fa0b626b48ed
subsection
3
30
Preliminaries
In this paper, we use f(x) as the one defined in (REF ), and F(x;\xi ) as a function whose stochastic nature comes from the random variable \xi representing a random index selected from the training set \lbrace 1,\ldots , n\rbrace . We use \Vert {x} \Vert to denote the \ell _2 norm of the vector x, and \langle {x, y} \...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.026250718161463737, -0.014834708534181118, -0.042184293270111084, 0.015842003747820854, -0.00494108721613884, 0.014300537295639515, 0.02452610805630684, 0.030463043600320816, 0.009424312971532345, 0.0096456129103899, -0.037514109164476395, 0.04282530024647713, 0.006421505939215422, 0.01...
22a4ee6aeac5883d64852278418ebd71242ca188
subsection
4
30
Stochastic Optimization Problems
In this paper, we consider the following stochastic optimization problem instead of the original deterministic version (REF ):\begin{split} \mathop {\min }_{x\in \mathbb {R}^{{d}}} &\quad \Psi (x):= \mathbb {E}_\xi [F(x; \xi )] + h(x), \end{split}where the stochastic nature comes from the random variable \xi , which in...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.02681722305715084, -0.015675177797675133, -0.018788844347000122, 0.003329258179292083, -0.026603540405631065, 0.04243134707212448, 0.012500458396971226, 0.0224367193877697, 0.027900900691747665, 0.03293771669268608, -0.04151556268334389, 0.012370722368359566, 0.006418122909963131, 0.021...
fb9969df762fc5b50c5e4502c42d2a85a45dabbd
subsection
5
30
Proximal Gradient Descent
The proximal operator is fundamental to many algorithms to solve problem (REF ) as well as its stochastic variant (REF ).Definition 1 (Proximal operator) The proximal operator \mathbf {prox}^{}_{} of a point x \in \mathbb {R}^{{d}} under a proper and closed function h with parameter \eta > 0 is defined as:\mathbf {pro...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03939890116453171, -0.0008120682323351502, -0.04281692951917648, -0.034515999257564545, 0.006172291003167629, 0.028961701318621635, 0.054719001054763794, -0.005779370199888945, 0.037018485367298126, 0.034241337329149246, -0.07470837235450745, 0.022262971848249435, -0.00500497268512845, ...
34a8e64c9b9a58e937eef70df167dacbe5f74d6a
subsection
6
30
Parallel Stochastic Optimization
Recent years have witnessed rapid development of parallel and distributed computation frameworks for large-scale machine learning problems. One popular architecture is called parameter server , , which consists of some worker nodes and server nodes. In this architecture, one or multiple master machines play the role of...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 249, "openalex_id": "https://openalex.org/W2168231600", "raw": "J. Dean, G. Corrado, R. Monga, K. Chen, M. Devin, M. Mao, A. Senior, P. Tucker, K. Yang, Q. V. Le, et al. Large scale distributed deep networks. In Advances in neural...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.04918117821216583, 0.0008504500146955252, -0.049760859459638596, -0.011395267210900784, 0.0007775134290568531, 0.0037374035455286503, -0.013500416651368141, -0.009412155486643314, 0.02663472108542919, 0.02388887293636799, -0.01584964245557785, 0.021859997883439064, 0.003500955644994974, ...
512dae4ebcc0e128ad16ece99d0fdf350ac4bc8d
subsection
7
30
Asynchronous Proximal Gradient Descent
We now present our asynchronous proximal gradient descent (Asyn-ProxSGD) algorithm, which is the main contribution in this paper. In the asynchronous algorithm, different workers may be in different local iterations due to random delays in computation and communication.For ease of presentation, let us first assume each...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.048718180507421494, 0.006978818215429783, -0.0179945919662714, -0.02150498703122139, 0.012278752401471138, 0.015445739030838013, 0.006570544093847275, -0.014873391948640347, 0.03427977114915848, 0.047252971678972244, -0.0218865517526865, 0.033913470804691315, -0.04096478596329689, 0.014...
19cf53c178bc3f57f8151b87f0245494c29475ad
subsection
8
30
Convergence Analysis
To facilitate the analysis of Algorithm REF , we rewrite it in an equivalent global view (from the server's perspective), as described in Algorithm . In this algorithm, we use an iteration counter k to keep track of how many times the model x has been updated on the server; k increments every time a push request (model...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.05467357859015465, -0.020807690918445587, -0.014164178632199764, -0.033255692571401596, 0.010060609318315983, -0.008870727382600307, 0.02274506352841854, -0.012821746990084648, 0.009778394363820553, 0.0333167128264904, -0.00720412889495492, 0.02661980874836445, -0.029228396713733673, 0....
6c43712e9292f327498d55f61768393a7c2ece49
subsection
9
30
Assumptions and Metrics
We make the following assumptions for convergence analysis. We assume that f(\cdot ) is a smooth function with the following properties:Assumption 1 (Lipschitz Gradient) For function f there are Lipschitz constants L>0 such that\Vert {\nabla f(x) - \nabla f(y)} \Vert \le L \Vert {x - y} \Vert , \forall x, y \in \mathb...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.1506.08272", "end": 985, "openalex_id": "https://openalex.org/W778657980", "raw": "X. Lian, Y. Huang, Y. Li, and J. Liu. Asynchronous parallel stochastic gradient for nonconvex optimization. In Advances in Neural Information ...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.008681871928274632, 0.02668645791709423, -0.05199968442320824, 0.021285081282258034, 0.03112657181918621, 0.0015792157500982285, 0.0386946015059948, -0.007129358127713203, 0.013694165274500847, 0.03057727962732315, -0.011901335790753365, 0.015822673216462135, 0.008399596437811852, 0.025...
c3b7d93bba80ca6c994c0c5483a6a309ee507542
subsection
10
30
Theoretical Results
We present our main convergence theorem as follows:Theorem 1 If Assumptions REF and REF hold and the step length sequence \lbrace \eta _k\rbrace in Algorithm  satisfies\eta _k \le \frac{1}{16L},\quad 6\eta _k L^2 T \sum _{l=1}^T \eta _{k+l} \le 1,for all k=1,2,\ldots , K, we have the following ergodic convergence rate ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/s10107-014-0846-1", "end": 1585, "openalex_id": "https://openalex.org/W2029463628", "raw": "S. Ghadimi, G. Lan, and H. Zhang. Mini-batch stochastic approximation methods for nonconvex stochastic composite optimization. Mathematical ...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.016843978315591812, -0.006030418910086155, -0.0213143453001976, -0.01585225947201252, 0.009489995427429676, 0.009116193279623985, -0.01678294874727726, 0.009161964990198612, 0.009070421569049358, 0.03445081785321236, -0.008360960520803928, 0.004844169598072767, -0.026333972811698914, 0....
ac12b6e57a151fc5398acf1878eba48dec2b091d
subsection
11
30
Theoretical Results
The reason is that by (REF ) and (REF ), as long as T is no more than O(K^{1/4}), the iteration complexity (from a global perspective) to achieve \epsilon -optimality is O(1/\epsilon ^2), which is independent from T.Remark 3 (Linear speedup w.r.t. number of workers) As the iteration complexity is O(1/\epsilon ^2) to a...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.1506.08272", "end": 1013, "openalex_id": "https://openalex.org/W778657980", "raw": "X. Lian, Y. Huang, Y. Li, and J. Liu. Asynchronous parallel stochastic gradient for nonconvex optimization. In Advances in Neural Information...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.04809332266449928, 0.011199397034943104, -0.029112329706549644, -0.007072099018841982, 0.014014504849910736, 0.008903062902390957, -0.03561225160956383, 0.0009698388166725636, 0.03625309094786644, 0.0343610942363739, -0.02595391683280468, 0.0053403121419250965, 0.0171347726136446, 0.009...
6e3928b4eb3e7d5938121b7aede1694ae41e08da
subsection
12
30
Experiments
We now present experimental results to confirm the capability and efficiency of our proposed algorithm to solve challenging non-convex non-smooth machine learning problems. We implemented our algorithm on TensorFlow , a flexible and efficient deep learning library. We execute our algorithm on Ray , a general-purpose fr...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.1605.08695", "end": 265, "openalex_id": "https://openalex.org/W2953384591", "raw": "M. Abadi, P. Barham, J. Chen, Z. Chen, A. Davis, J. Dean, et al. Tensorflow: A system for large-scale machine learning. In 12th USENIX Sympos...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03455529734492302, 0.0027511450462043285, -0.02074233628809452, 0.01967393048107624, -0.014087693765759468, 0.017674485221505165, 0.019536564126610756, 0.01709449291229248, 0.021078120917081833, 0.06953795254230499, -0.0014995838282629848, -0.015781879425048828, -0.01576661691069603, 0....
f6ca96f84d3c6435be26b43c4330145da37e282b
subsection
13
30
Experiments
For all experiments, we evaluate function suboptimality, which is the gap f(x) - f(\hat{x}), against the number of sample gradients processed by the server (divided by the total number of samples n), and then against time. [Figure: Performance of ProxGD and Async-ProxSGD on a9a (left) and mnist (right) datasets. Here t...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.1506.08272", "end": 1013, "openalex_id": "https://openalex.org/W778657980", "raw": "X. Lian, Y. Huang, Y. Li, and J. Liu. Asynchronous parallel stochastic gradient for nonconvex optimization. In Advances in Neural Information...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.061034850776195526, -0.017410190775990486, -0.031829673796892166, -0.0038795277941972017, 0.0027656417805701494, 0.008171040564775467, -0.025497308000922203, 0.004436470568180084, 0.0120620122179389, 0.03314192220568657, -0.02941879816353321, 0.027709823101758957, -0.022537117823958397, ...
e17182096ffed9b43c0949751e47ec0c064a2a93
subsection
14
30
Related Work
Stochastic optimization problems have been studied since the seminal work in 1951 , in which a classical stochastic approximation algorithm is proposed for solving a class of strongly convex problems. Since then, a series of studies on stochastic programming have focused on convex problems using SGD , , . The convergen...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1214/aoms/1177729586", "end": 200, "openalex_id": "https://openalex.org/W1994616650", "raw": "H. Robbins and S. Monro. A stochastic approximation method. The annals of mathematical statistics, pages 400–407, 1951.", "source_ref_id"...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.056448448449373245, 0.003644357668235898, -0.05007129907608032, -0.04253466799855232, -0.01673620194196701, 0.04009365662932396, 0.03585239127278328, -0.012975514866411686, 0.03185523301362991, 0.05205462500452995, -0.031702667474746704, 0.007521374151110649, 0.009519954212009907, 0.036...
98ea5604e354bba5e2063e37ddd632f2725a5160
subsection
15
30
Concluding Remarks
In this paper, we study asynchronous parallel implementations of stochastic proximal gradient methods for solving nonconvex optimization problems, with convex yet possibly nonsmooth regularization. However, compared to asynchronous parallel stochastic gradient descent (Asyn-SGD), which is targeting smooth optimization,...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.043796274811029434, 0.012955762445926666, -0.01707596890628338, -0.036288343369960785, -0.023393619805574417, 0.03406037762761116, 0.037753306329250336, 0.016053548082709312, 0.033907778561115265, 0.07434684783220291, -0.03229021653532982, 0.02751382626593113, -0.0059437802992761135, 0....
75d4480eedcebc48a041673b1a6161843f96dcab
subsection
16
30
Auxiliary Lemmas
Lemma 1 () For all y \leftarrow \mathbf {prox}^{}_{\eta h}(x - \eta g), we have:\langle {g, y-x} \rangle + (h(y) - h(x)) \le -\frac{\Vert {y-x} \Vert _2^2}{\eta }.Due to slightly different notations and definitions in , we provide a proof here for completeness. We refer readers to for more details. By the definition ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/s10107-014-0846-1", "end": 165, "openalex_id": "https://openalex.org/W2029463628", "raw": "S. Ghadimi, G. Lan, and H. Zhang. Mini-batch stochastic approximation methods for nonconvex stochastic composite optimization. Mathematical P...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.05284871533513069, 0.02700410597026348, -0.019177492707967758, -0.029078997671604156, 0.009939647279679775, -0.025112293660640717, 0.015203159302473068, 0.016492338851094246, 0.021069305017590523, -0.0020443787798285484, -0.031001325696706772, 0.016339773312211037, 0.0035490565933287144, ...
7f7f72b99027c5a099d99ee9513b8be5feb7df63
subsection
17
30
Auxiliary Lemmas
Let z substitute u in the first inequality and y in the second one, we have\langle {G + \frac{y-x}{\eta } + p, z-y} \rangle &\ge 0, \\ \langle {g + \frac{z-x}{\eta } + q, y-z} \rangle &\ge 0.Then, we have\langle {G, z-y} \rangle &\ge \langle {\frac{y-x}{\eta }, y-z} \rangle + \langle {p, y-z} \rangle , \\ &= \frac{1}{\...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/s10107-014-0846-1", "end": 1049, "openalex_id": "https://openalex.org/W2029463628", "raw": "S. Ghadimi, G. Lan, and H. Zhang. Mini-batch stochastic approximation methods for nonconvex stochastic composite optimization. Mathematical ...
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.038845282047986984, 0.02331327274441719, -0.011465919204056263, 0.0015867672627791762, 0.030453726649284363, -0.010138527490198612, 0.0366177074611187, 0.013647723942995071, 0.023786252364516258, -0.0006861052243039012, -0.021406101062893867, 0.025540851056575775, 0.0029313236009329557, ...
ee52efcf6515b5dbfec35174ba45bf5a94f4c754
subsection
18
30
Auxiliary Lemmas
We denote \tilde{G}_k as the average of delayed stochastic gradients and \tilde{g}_k as the average of delayed true gradients, respectively:\tilde{G}_k &:= \frac{1}{N}\sum _{i=1}^N \nabla F(x_{t(k,i)}; \xi _{t(k,i), i}) \\ \tilde{g}_k &:= \frac{1}{N}\sum _{i=1}^N \nabla f(x_{t(k,i)}).Moreover, we denote \delta _k := \t...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.02342299558222294, -0.005874822847545147, -0.016739429906010628, -0.03146463632583618, -0.010002458468079567, -0.004940191749483347, 0.029709817841649055, 0.034913234412670135, -0.00768304755911231, 0.02703944407403469, -0.014770982787013054, 0.0005750841228291392, -0.037110570818185806, ...
762a70e61afe229b268bfbf45b8a27dce0b04bd9
subsection
19
30
Milestone lemmas
We put some key results of convergence analysis as milestone lemmas listed below, and the detailed proof is listed in REF .Lemma 5 (Decent Lemma)\mathbb {E}[\Psi (x_{k+1}) \le \mathbb {E}[\Psi (x_k)|\mathcal {F}_k] - \frac{\eta _k - 4L\eta _k^2}{2} \Vert {P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 + \frac{\eta _k}{2}\Vert ...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ 0.002553399885073304, 0.03414195030927658, -0.014904685318470001, -0.0024466109462082386, 0.0006822096765972674, -0.0024256345350295305, 0.027063367888331413, 0.009199104271829128, 0.015240306966006756, 0.05168584734201431, 0.013554567471146584, -0.00004016503589809872, -0.010953494347631931...
1e14f7509e92aea948c3bbdb4dd4545495b4c0b6
subsection
20
30
Proof of Theorem
From the fact 2\Vert {a} \Vert ^2 + 2\Vert {b} \Vert ^2 \ge \Vert {a+b} \Vert ^2, we have\Vert {P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 + \Vert {g_k - \tilde{g}_k} \Vert ^2 &\ge \Vert {P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 + \Vert {P(x_k, g_k, \eta _k) - P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 \\ &\ge \frac{1}{2} \Vert ...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.05426287651062012, 0.03668390214443207, 0.0019312838558107615, -0.013611497357487679, -0.004371854476630688, -0.004524449817836285, 0.02455257810652256, 0.04614480957388878, 0.035737812519073486, -0.0035497474018484354, 0.0055620972998440266, -0.01709067076444626, -0.024155830964446068, ...
62167ac07a3092847d862538ea5d5f7597b189e3
subsection
21
30
Proof of Theorem
According to our condition of \eta \le \frac{1}{16L}, we have 8L\eta _k^2 - \eta < 0 and therefore\begin{split} &\quad \ \mathbb {E}[\Psi (x_{k+1})|\mathcal {F}_k] \\ &\le \mathbb {E}[\Psi (x_k)|\mathcal {F}_k] + \frac{\eta _k}{2}\Vert {g_k-\tilde{g}_k} \Vert ^2 + \frac{4L\eta _k^2-\eta _k}{2} \Vert {P(x_k, \tilde{g}_k...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03362242877483368, 0.022608183324337006, -0.016521364450454712, -0.0328291580080986, -0.006956364028155804, 0.004328658804297447, 0.03575815260410309, 0.033683449029922485, 0.016734937205910683, 0.0012871562503278255, 0.00835984107106924, 0.02840515412390232, -0.03046460449695587, 0.002...
68159042781862e91ad64086a87d8c9a2a1d21d2
subsection
22
30
Proof of Theorem
\end{split}Apply Lemma REF we have\begin{split} &\quad \ \mathbb {E}[\Psi (x_{k+1})|\mathcal {F}_k] \\ &\le \mathbb {E}[\Psi (x_k)|\mathcal {F}_k] - \frac{\eta _k-8L\eta _k^2}{8} \Vert {P(x_k, g_k, \eta _k)} \Vert ^2 + \frac{L\eta _k^2}{N}\sigma ^2 - \frac{\eta _k}{4}\Vert {P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 \\ &\q...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.048083510249853134, 0.01949578896164894, -0.004698515869677067, -0.02931995876133442, -0.00470614293590188, 0.009923325851559639, 0.04347652569413185, 0.02855721302330494, -0.0008614263497292995, 0.0075931367464363575, -0.013584507629275322, 0.04387315362691879, -0.03706945851445198, 0....
fb3a04fa60f5ef5bc42aa225e65cf2ec04534125
subsection
23
30
Proof of Theorem
\end{split}By taking telescope sum, we have&\quad \ \mathbb {E}[\Psi (x_{K+1})|\mathcal {F}_K] \\ &\le \Psi (x_1) - \sum _{k=1}^K \frac{\eta _k-8L\eta _k^2}{8} \Vert {P(x_k, g_k, \eta _k)} \Vert ^2 - \sum _{k=1}^K \left( \frac{\eta _k}{4}- \frac{3\eta _k^2L^2T}{2}\sum _{l=1}^{l_k}\eta _{k+l} \right) \Vert { P(x_{k}, \t...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.019690338522195816, 0.04047967121005058, -0.009120137430727482, -0.03718268498778343, 0.007509803399443626, 0.015859117731451988, 0.011356288567185402, 0.011432607658207417, -0.018026581034064293, 0.014241151511669159, -0.015874382108449936, 0.03553419187664986, -0.04371560364961624, 0....
2eb327321ce67a23f19150a27dd6d5c8c147155a
subsection
24
30
Proof of Corollary
From the condition of Corollary, we have\eta \le \frac{1}{16L(T+1)^2}.It is clear that the above inequality also satisfies the condition in Theorem REF . By doing so, we can have Furthermore, we have\frac{3LT^2\eta }{2} &\le \frac{3LT^2}{2}\cdot \frac{1}{16L(T+1)^2} \le 1, \\ \frac{3L^2T^2\eta ^3}{2} &\le L\eta ^2.Sinc...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03275377303361893, 0.02988438308238983, -0.008363964036107063, -0.03217379003763199, -0.008119761012494564, 0.03385268524289131, -0.024282969534397125, 0.01913943514227867, -0.0066507249139249325, -0.012019382789731026, 0.0009601197089068592, 0.02008572407066822, -0.04554392024874687, 0...
218ee003b70f108157d2af338c536e088d4f9f1e
subsection
25
30
Proof of milestone lemmas
[Proof of Lemma REF ] Let \bar{x}_{k+1} = \mathbf {prox}^{}_{\eta _k h}(x_k - \eta _k \tilde{g}_k) and apply Lemma REF , we have\begin{split} \Psi (x_{k+1}) &\le \Psi (\bar{x}_{k+1}) + \langle {x_{k+1}-\bar{x}_{k+1}, \nabla f(x_k) - \tilde{G}_k} \rangle + \left( \frac{L}{2} - \frac{1}{2\eta _k} \right) \Vert {x_{k+1}-x...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.028784381225705147, 0.037636417895555496, -0.017078325152397156, -0.009676191955804825, -0.012125764042139053, -0.010698754340410233, 0.05210492014884949, 0.0034072711132466793, 0.026266129687428474, 0.0038632273208349943, -0.038307953625917435, 0.03064636141061783, -0.024968847632408142,...
27a927c428f700c1c592147d8e80845381ed965b
subsection
26
30
Proof of milestone lemmas
\end{split}Now we turn to bound \Psi (\bar{x}_{k+1}) as follows:\begin{split} f(\bar{x}_{k+1}) &\le f(x_k) + \langle {\nabla f(x_k), \bar{x}_{k+1} - x_k} \rangle + \frac{L}{2}\Vert {\bar{x}_{k+1} - x_k} \Vert ^2 \\ &= f(x_k) + \langle {g_k, \bar{x}_{k+1} - x_k} \rangle + \frac{\eta _k^2 L}{2}\Vert {P(x_k, \tilde{g}_k, ...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.021085213869810104, 0.024594329297542572, -0.0020883057732135057, -0.00809385348111391, -0.01113763079047203, 0.0009850319474935532, 0.03933262079954147, -0.006407951936125755, 0.023129655048251152, 0.0230838842689991, -0.039546217769384384, 0.022122690454125404, -0.017469298094511032, ...
80de981796a3cbcc57edc007f6699a214252f14d
subsection
27
30
Proof of milestone lemmas
By rearranging terms on both sides, we have\Psi (\bar{x}_{k+1}) \le \Psi (x_k) - (\eta _k - \frac{\eta _k^2 L}{2}) \Vert {P(x_k, \tilde{g}_k, \eta _k)} \Vert ^2 + \langle {g_k - \tilde{g}_k, \bar{x}_{k+1} - x_k} \rangleTaking the summation of (REF ) and (REF ), we have\begin{split} &\quad \ \Psi (x_{k+1}) \\ &\le \Psi ...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.03872823715209961, 0.025971416383981705, -0.00946079846471548, 0.005630700848996639, 0.005665034521371126, 0.005062290001660585, 0.03622570261359215, 0.0015373796923086047, 0.029358994215726852, 0.025818822905421257, -0.01701417751610279, 0.01571713201701641, -0.021164720878005028, 0.02...
43d1c3f3bf441b6dd114b288e52d143f0771139c
subsection
28
30
Proof of milestone lemmas
\end{split}Therefore, we have\begin{split} &\quad \ \mathbb {E}[\Psi (x_{k+1})|\mathcal {F}_k] \\ &\le \mathbb {E}[\Psi (x_k)|\mathcal {F}_k] + \mathbb {E}[\langle {x_{k+1}-x_k, g_k-\tilde{g}_k} \rangle |\mathcal {F}_k] + \frac{L\eta _k^2 - \eta _k}{2} \mathbb {E}[\Vert {P(x_k, \tilde{G}_k, \eta _k)} \Vert ^2|\mathcal ...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.04210733622312546, 0.008818130008876324, -0.005026944447308779, -0.04152759537100792, -0.0003537550219334662, 0.0031733063515275717, 0.041619133204221725, 0.020153546705842018, 0.00824601948261261, 0.032953567802906036, -0.007528974674642086, 0.013860330916941166, -0.0324348546564579, 0...
fdd0256d8c52c5888aa38a9eb57e0ab8d57bf097
subsection
29
30
Proof of milestone lemmas
By taking the expectation on both sides, we have\begin{split} \mathbb {E}[\Vert {x_k - x_{k-\tau }} \Vert ^2 ] &\le 2\tau \sum _{l=1}^{\tau } \eta _{k-l}^2\Vert {\tilde{G}_{k-l}-\tilde{g}_{k-l}} \Vert ^2 + 2\left\Vert {\sum _{l=1}^{\tau } \eta _{k-l} P(x_{k-l}, \tilde{g}_{k-l}, \eta _{k-l}) } \right\Vert ^2 \\ &\le \fr...
{ "cite_spans": [] }
1802.08880
Asynchronous Stochastic Proximal Methods for Nonconvex Nonsmooth Optimization
[ "Rui Zhu", "Di Niu", "Zongpeng Li" ]
[ "cs.LG" ]
2,018
en
Computer Science
[ -0.051929082721471786, 0.03228024020791054, -0.004553710576146841, -0.04387427866458893, -0.010762622579932213, 0.023767776787281036, 0.019526800140738487, 0.041708022356033325, 0.009351505897939205, 0.01977088488638401, 0.004138003569096327, 0.011929653584957123, -0.04579644650220871, 0.0...
967d0ef79da3c987aed954aaa3642bacb7fd5acf
abstract
0
12
Abstract
Semantic parsing offers many opportunities to improve natural language understanding. We present a semantically annotated parallel corpus for English, German, Italian, and Dutch where sentences are aligned with scoped meaning representations in order to capture the semantics of negation, modals, quantification, and pre...
{ "cite_spans": [] }
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.05300932377576828, -0.006698645185679197, 0.020309560000896454, -0.0016155331395566463, -0.018890485167503357, -0.03387469798326492, 0.03028886206448078, -0.005851777736097574, 0.011215271428227425, 0.03018205054104328, -0.0073090000078082085, 0.01332862488925457, -0.0073280734941363335, ...
88ebff6f888d384f46aa3cb658938b32eb007c83
subsection
1
12
Introduction
Semantic parsing is the task of assigning meaning representations to natural language expressions. Informally speaking, a meaning representation describes who did what to whom, when, and where, and to what extent this is the case or not. The availability of open-domain, wide coverage semantic parsers has the potential ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 1174, "openalex_id": "https://openalex.org/W2250375035", "raw": "Cai, S. and Knight, K. (2013). Smatch: an evaluation metric for semantic feature structures. In Proceedings of the 51st Annual Meeting of the Association for Computa...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.05927475914359093, -0.01200298685580492, -0.0003083252813667059, 0.013857230544090271, -0.018710313364863396, -0.006993475370109081, 0.027973901480436325, -0.012117446400225163, 0.0171689260751009, 0.024997955188155174, -0.01097285095602274, 0.003141913330182433, 0.0006686342530883849, ...
3a20de804c222fb77478479d7d532515f67edef7
subsection
2
12
Discourse Representation Structures
The backbone of the meaning representations in our annotated corpus is formed by the Discourse Representation Structures (DRS) of Discourse Representation Theory . Our version of DRS integrates WordNet senses , adopts a neo-Davidsonian analysis of events employing VerbNet roles , and includes an extensive set of compar...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 163, "openalex_id": "", "raw": "Kamp, H. and Reyle, U. (1993). From Discourse to Logic; An Introduction to Modeltheoretic Semantics of Natural Language, Formal Logic and DRT. Kluwer, Dordrecht.", "source_ref_id": "d3db9f3e70...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.031404174864292145, -0.0004787686630152166, -0.009888194501399994, -0.00489831855520606, -0.02882530726492405, -0.014481321908533573, 0.03320480138063431, 0.008858174085617065, 0.03302168846130371, 0.03412037342786789, -0.004127711057662964, -0.024140622466802597, 0.0004699467390310019, ...
5ea2de15a9957b4fb5554ec3fb54f9e1f2e71c6c
subsection
3
12
Discourse Representation Structures
The resulting clauses are then of the form t R t' or t R t' t” where R \in \cal {C}\cup \cal {T}\cup \cal {O}. The result of translating DRSs to sets of clauses is shown in Figure REF . In a clausal form, it is assumed that different variables are represented with different variable names and vice versa. Due to this, b...
{ "cite_spans": [] }
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.022008730098605156, 0.030555810779333115, -0.02591596730053425, 0.026358583942055702, -0.03632508963346481, -0.0054449476301670074, 0.03023529425263405, 0.010767794214189053, 0.007417644374072552, 0.016575230285525322, -0.01199643686413765, 0.009180479682981968, 0.00009092003892874345, ...
ebbd46f5236b51ce7c11861addc4fc02d713fde8
subsection
4
12
Comparing DRSs to AMRs
Since DRSs in a clausal form come close to the triple notation of AMRs , and both aim to model meaning of natural language expressions, it is instructive to compare these two meaning representations. The main difference between AMRs and DRSs is that the latter ones have explicit scopes (boxes) and scopal operators such...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 199, "openalex_id": "https://openalex.org/W2250375035", "raw": "Cai, S. and Knight, K. (2013). Smatch: an evaluation metric for semantic feature structures. In Proceedings of the 51st Annual Meeting of the Association for Computat...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.022152448073029518, -0.009809933602809906, -0.00825376994907856, -0.027431199327111244, -0.038293831050395966, -0.01797216385602951, 0.025646187365055084, -0.012083153240382671, 0.014547078870236874, 0.01977243274450302, -0.04177231341600418, -0.005191026255488396, -0.015424327924847603, ...
a1810e842deeccaa8493f5e6fc0060fb68a423ed
subsection
5
12
The Parallel Meaning Bank
The scoped meaning representations, integrating word senses, thematic roles, and the list of operators, form the final product of our semantically annotated corpus: the Parallel Meaning Bank. The PMB is a semantically annotated corpus of English texts aligned with translations in Dutch, German and Italian . It uses the...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.18653/v1/e17-2039", "end": 308, "openalex_id": "https://openalex.org/W2594470997", "raw": "Abzianidze, L., Bjerva, J., Evang, K., Haagsma, H., van Noord, R., Ludmann, P., Nguyen, D.-D., and Bos, J. (2017). The Parallel Meaning Bank: Towa...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.03128596395254135, -0.043739303946495056, -0.0285236407071352, 0.011911557987332344, 0.0215644221752882, -0.02267850749194622, 0.02463197335600853, -0.01823742687702179, 0.03534550592303276, 0.029866648837924004, -0.026768574491143227, 0.0051087685860693455, -0.020358767360448837, 0.019...
21b369ec3ad28babb5229d727eaa083f76ab1dae
subsection
6
12
Evaluation by Matching
In the context of the Parallel Meaning Bank there are two main reasons to verify whether two scoped meaning representations capture the same meaning or not: (1) to be able to evaluate semantic parsers that produce scoped meaning representations by comparing gold-standard DRSs to system output; and (2) to check whether ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 699, "openalex_id": "", "raw": "Blackburn, P. and Bos, J. (2005). Representation and Inference for Natural Language. A First Course in Computational Semantics. CSLI.", "source_ref_id": "fcaa460d4a524ff0e0bddf25690434c4dab9fd...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.02890588343143463, -0.007642355281859636, -0.0195198655128479, 0.002678449032828212, -0.01092746201902628, 0.011301376856863499, 0.06501534581184387, 0.0006600736523978412, 0.009721778333187103, 0.046914830803871155, -0.05463730916380882, -0.0016606765566393733, -0.0045938072726130486, ...
823ca944b000d39fa7238c1de304bb338bb1efe2
subsection
7
12
Evaluation by Matching
Before matching two DRSs, redundant REF-clauses are removed. A REF-clause \langle [0]b1 REF x1\rangle is redundant if its discourse referent [0]x1 occurs in some basic condition of the same DRS [0]b1. Figure REF shows some examples of redundant REF-clauses. Not removing these redundant clauses would lead to inflated ma...
{ "cite_spans": [] }
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.059730835258960724, 0.017107374966144562, -0.0035195257514715195, 0.004799526650458574, -0.012536760419607162, -0.01985432207584381, 0.05399277061223984, 0.025439782068133354, 0.014368059113621712, 0.012193392030894756, -0.04517201706767082, -0.037083785980939865, -0.004444712772965431, ...
7cf2f4364bc71b00c8aa6e8c9ff05f3011cfeb14
subsection
8
12
Evaluating Matching
As we showed in Figure REF , DRSs are about twice as large as AMRs. This increase in size might be problematic, since it increases the average runtime for comparing DRSs. Moreover, if there are more variables, more restarts might be needed to ensure a reliable score, again increasing runtime.Therefore, our goal is that...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 957, "openalex_id": "", "raw": "van Noord, R. and Bos, J. (2017). Neural semantic parsing by character-based translation: Experiments with Abstract Meaning Representations. Computational Linguistics in the Netherlands Journal, 7:9...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ 0.008657706901431084, -0.010915576480329037, -0.0329221710562706, -0.016628900542855263, -0.01635429449379444, 0.01989365741610527, -0.00016090177814476192, -0.004569133743643761, 0.017788346856832504, 0.015149081125855446, -0.04521840438246727, 0.019191887229681015, -0.030465975403785706, ...
76c5066f13a22d285e1c86512d4caab1dd4f79be
subsection
9
12
Semantic Parsing
The first purpose of counter is to evaluate semantic parsers for DRSs. Since this is a new task, there are no existing systems that are able to do this. Therefore, we show the results of three baseline systems pmb pipeline, Spar, and amr2drs (Subsection REF ).Spar and amr2drs are available at: https://github.com/RikVN/...
{ "cite_spans": [] }
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.04306363686919212, -0.01756422594189644, -0.04101880267262459, -0.007588020525872707, -0.0016566605772823095, 0.005485959351062775, 0.02534681186079979, 0.007610910106450319, 0.0016852730186656117, 0.04947282373905182, -0.06006324291229248, -0.0062222531996667385, -0.03213749825954437, ...
f82c345189736c405bdcf82974452d1f11e0b27f
subsection
10
12
Comparing Translations
The second purpose of counter is checking whether translations are meaning-preserving. As a pilot study, we compare the gold standard meaning representations of German, Italian and Dutch translations in the release to their English counterparts. The results are shown in Table REF . The high F-scores indicate that the m...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 1427, "openalex_id": "https://openalex.org/W86887328", "raw": "Cucerzan, S. (2007). Large-scale named entity disambiguation based on Wikipedia data. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Langu...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.026490652933716774, -0.021241353824734688, -0.03268604725599289, -0.006843927316367626, 0.006576884537935257, -0.03570744767785072, -0.0006079993909224868, 0.01785372383892536, -0.02385074459016323, 0.0366840623319149, 0.008118103258311749, 0.014221941120922565, -0.05017354339361191, 0....
958d2e056db865b5417a4584601025b1425ee8f1
subsection
11
12
Conclusions and Future Work
Large semantically annotated corpora are rare. Within the Parallel Meaning Bank project, we are creating a large, open-domain corpus annotated with formal meaning representations. We take advantage of parallel corpora, enabling the production of meaning representations for several languages at the same time. Currently,...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 1641, "openalex_id": "https://openalex.org/W3089269909", "raw": "Hamp, B. and Feldweg, H. (1997). GermaNet - a lexical-semantic net for German. In In Proceedings of ACL workshop Automatic Information Extraction and Building of Lex...
1802.08599
Evaluating Scoped Meaning Representations
[ "Rik van Noord", "Lasha Abzianidze", "Hessel Haagsma", "Johan Bos" ]
[ "cs.CL" ]
2,018
en
Computer Science
[ -0.03122611716389656, 0.021977325901389122, 0.016681401059031487, 0.02472449280321598, -0.04197058826684952, -0.014605764299631119, 0.03510267287492752, -0.009347994811832905, 0.03110402077436447, 0.017154524102807045, -0.001487094210460782, 0.02588440664112568, -0.009271684102714062, 0.03...
cdb1fc34cb9d352f4feef85cdf80c7b0cf3cb34b
abstract
0
33
Abstract
Citation parsing is fundamental for search engines within academia and the protection of intellectual property. Meticulous extraction is further needed when evaluating the similarity of documents and calculating their citation impact. Citation parsing involves the identification and dissection of citation strings into ...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.024663250893354416, -0.02009993977844715, -0.01154563669115305, 0.045480500906705856, 0.003342358861118555, -0.018009057268500328, 0.036140210926532745, -0.010965684428811073, 0.03162268176674843, 0.013934126123785973, -0.03369830548763275, -0.03256892040371895, -0.07264670729637146, -0...
4ba8d7d66b53a6830a4f9923ea97cedfb0e539c2
subsection
1
33
Introduction
In the past decade there has been an explosion in the amount of scientific publications accessible on many different library websites. In order to search and use these libraries effectively there is a amplifying demand for accurate organization. Organizing these articles by using the meta-data associated with them, all...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 421, "openalex_id": "", "raw": "Joeran Beel, Bela Gipp, Stefan Langer, and Corinna Breitinger. Research paper recommender systems: A literature survey. International Journal on Digital Libraries, (4):305–338, 2016. doi: 10.1007/s0...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.04771709069609642, -0.00913763977587223, -0.05787680670619011, 0.04905951768159866, 0.028114641085267067, 0.00130428746342659, -0.00450017349794507, 0.009633421897888184, 0.00913763977587223, 0.040120188146829605, -0.04307962581515312, 0.0037259908858686686, -0.03194360435009003, 0.0187...
9f006e5c546f2ca7b4b818db2117d7f1786b18c3
subsection
2
33
Introduction
Argon, Cenk, and Steven W. McLaughlin 2002 A parallel decoder for low latency decoding of turbo product codes. IEEE Communications Letters 6(2). IEEE Communications Letters: 70–72. Retrieved. from https://doi.org/10.1109/4234.984698. ARGON, Cenk; MCLAUGHLIN, Steven W. A parallel decoder for low latency decoding of turb...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.027239592745900154, 0.009530042298138142, -0.015840167179703712, 0.0036739115603268147, 0.02582038752734661, 0.032199181616306305, 0.003385873744264245, -0.003719692351296544, -0.027788963168859482, 0.043858032673597336, -0.026400277391076088, -0.012833892367780209, -0.012292152270674706,...
29c2742d7202e04a23aba4eee605a24b0061a382
subsection
3
33
Motivation
Accurate citation parsing is a necessity within academic search engines and for the security of intellectual property. ‘Automated extraction of bibliographic data, such as article titles, author names, abstracts, and references is essential to the affordable creation of large citation databases’ [1]. It aids in the ide...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/978-3-642-40501-3_37", "end": 768, "openalex_id": "https://openalex.org/W1761420992", "raw": "Mateusz Fedoryszak, Dominika Tkaczyk, and Lukasz Bolikowski. Large scale citation matching using apache hadoop. In Trond Aalberg, Christos...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.029715456068515778, -0.0011650503147393465, -0.04426809772849083, 0.04991220310330391, 0.012935681268572807, -0.005072068423032761, 0.044237587600946426, -0.026115432381629944, 0.036671437323093414, 0.02916629984974861, -0.024437453597784042, -0.008473786525428295, -0.04304775223135948, ...
226cc35a87b0695867c41f2ba9ef1e65f37e0533
subsection
4
33
Research Problem
Over the years many approaches to reference parsing have been proposed, including regular expressions, knowledge-based approaches and supervised machine learning. Machine learning-based solutions, in particular those falling into the category of supervised sequence tagging, are considered a state-of-the-art technique f...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 709, "openalex_id": "", "raw": "Andrew Kachites McCallum, Kamal Nigam, Jason Rennie, and Kristie Seymore. Automating the construction of internet portals with machine learning. Information Retrieval, 3(2):127–163, 2000.", "s...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.07190237194299698, 0.0006194369634613395, -0.0471821166574955, 0.011421672999858856, -0.0010681591229513288, 0.008148528635501862, 0.02645982801914215, 0.005466686096042395, -0.006290694698691368, 0.010246697813272476, -0.0034352762158960104, -0.013817401602864265, 0.0056497990153729916, ...
a00d4ae5811501ae4671821ee0c8661bb46e18f6
subsection
5
33
Research Problem
Rule-based methods have some draw-backs and only really have success in specific small to moderate journal areas due to the fact that journal publishers regularly are given specifics on what predefined citation styles to use. These rule-based methods are also less adaptable and are difficult to optimize under these cir...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.060913242399692535, 0.016403241083025932, -0.03005989082157612, 0.032074056565761566, -0.006885546259582043, 0.007698078639805317, -0.002969749504700303, 0.03726205974817276, -0.0025024479255080223, 0.01905827596783638, -0.05700698122382164, 0.003452309872955084, -0.006946581415832043, ...
4412542e534c2b329a1151e6f39194ee6402deed
subsection
6
33
Research Goal
The research goal is to build a massive, open source data set which can be used for training high end citation parsing machine learning tools and evaluate their performance. This goal can be broken into gathering BibTeX entries, citation generation into thousands of different styles and annotating of citations. The Bib...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 818, "openalex_id": "", "raw": "Michel Krämer. citeproc-java: A citation style language (csl) processor for java. URL http://michel-kraemer.github.io/citeproc-java/.", "source_ref_id": "0cffb25ebf229bc6fc2a4719f333cbca3b237d...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.05780927836894989, -0.027348000556230545, -0.06684388965368271, 0.03885491564869881, 0.01270491722971201, -0.012086839415133, 0.04007580876350403, 0.0018141705077141523, 0.04041155427694321, 0.03696253150701523, -0.020022643730044365, -0.014101313427090645, -0.0288130734115839, 0.020785...
ff9648b98ee4efddbe502742d886b6a6ec4df8fd
subsection
7
33
State of the Art
Numerous methodology have been used in research articles to solve the generalized problem of meta data extraction in reference strings. One of which is "A structural SVM approach for reference parsing" [1]. This approach involves using structural support vector machines which is a supervised machine learning algorithm/...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1186/1471-2105-12-s3-s7", "end": 561, "openalex_id": "https://openalex.org/W2130371208", "raw": "Xiaoli Zhang, Jie Zou, Daniel X Le, and George R Thoma. A structural svm approach for reference parsing. BMC bioinformatics, 12(3):S7, 2011a...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.056782353669404984, -0.024348370730876923, -0.02622484415769577, 0.041343413293361664, 0.011647858656942844, -0.01810871995985508, 0.027140196412801743, 0.005858254618942738, 0.00867296289652586, 0.026423171162605286, -0.04826957732439041, 0.01673569157719612, 0.006598164793103933, 0.05...
3c131b58d28c2caa4ef8605898b2057b1343a176
subsection
8
33
State of the Art
The extraction process proceeds by determining the arrangement of states that is most probabilistic for generating the complete reference and then placing corresponding labels with their fields according to the sequence of the states. Yin et al uses a "dynamic programming solution called the viterbi algorithm" which so...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/978-3-540-30544-6_33", "end": 354, "openalex_id": "https://openalex.org/W2158789913", "raw": "Ping Yin, Ming Zhang, ZhiHong Deng, and DongQing Yang. Metadata extraction from bibliographies using bigram hmm. In International Conferen...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.05411865562200546, -0.0057193925604224205, -0.040718719363212585, 0.0452972836792469, 0.004212281201034784, -0.013438092544674873, 0.026143614202737808, -0.02528894878923893, 0.025258425623178482, 0.030859537422657013, -0.0644051656126976, -0.010454393923282623, -0.005330214276909828, 0...
46cd720e18df6bba3ec3bb0f7c705be1d885df50
subsection
9
33
Pre-Processing of documents
Pre-Processing is important because of the crucial need for accuracy when calculating citation density, connections and making recommendations. It is needed because of errors obtained in a lot of databases because of human error upon citing and creating references and also through formatting issues which do not maintai...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 1359, "openalex_id": "https://openalex.org/W1559499673", "raw": "Isaac G Councill, C Lee Giles, and Min-Yen Kan. Parscit: an open-source crf reference string parsing package. In LREC, volume 8, pages 661–667, 2008.", "source...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.07177808880805969, -0.015854045748710632, -0.06051698327064514, 0.020080775022506714, 0.02092001587152481, 0.008056724444031715, 0.008056724444031715, -0.02600124664604664, -0.008117759600281715, 0.034149523824453354, -0.03744545578956604, -0.001869221101514995, -0.007042004261165857, 0...
450d9e64fa094784bcdddf31724f38be788544bd
subsection
10
33
Pre-Processing of documents
We just have more data.". Of course this is not always necessarily true, but in general more data is rarely a negative impact on results. A paper describing the means by which human complexities cannot be extrapolated through mathematical equations and instead accept those complexities, use what is useful and available...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1109/mis.2009.36", "end": 348, "openalex_id": "https://openalex.org/W2103018059", "raw": "Alon Halevy, Peter Norvig, and Fernando Pereira. The unreasonable effectiveness of data. IEEE Intelligent Systems, 24(2):8–12, 2009.", "sourc...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.051598772406578064, -0.0037417500279843807, -0.05641994625329971, 0.0329853780567646, -0.025890927761793137, -0.00814717449247837, 0.022122478112578392, 0.020810386165976524, 0.002166477032005787, -0.011839339509606361, -0.02689788118004799, 0.03847785294055939, -0.024746662005782127, 0...
c0239be062732e3e97637318c690c27ce7bf4b35
subsection
11
33
Introduction to Scraping
After trudging through the internet looking for data-sets that match my criteria and coming up empty. Automated scraping seemed like the most effective solution considering there were many sites that offer articles in BibTeX format with the click of a button. Scraping involves the requesting of html pages/files from se...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.019331710413098335, -0.021742451936006546, -0.039304252713918686, 0.015013735741376877, -0.009704000316560268, -0.02473299391567707, 0.002691105240955949, 0.001827128930017352, -0.011443396098911762, 0.056362539529800415, -0.03271285444498062, -0.003225130494683981, -0.024458352476358414,...
9ee7e7dbe89a12f2c098b6dfb6e3f552e6291030
subsection
12
33
Security and Reliable Sources without inaccuracies
All of the data scraped was freely available to the public and I don't need to consider data protection laws when porting it to a database. A main concept behind this investigation was making sure the data collected was of good standard from reliable sources so that the data did not contain errors or artificial entries...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.025379620492458344, -0.027623038738965988, -0.05579548329114914, 0.024921780452132225, 0.0020507434383034706, -0.034887440502643585, 0.049782514572143555, -0.02281571365892887, 0.027363594621419907, 0.018679888918995857, -0.019977103918790817, -0.014483017846941948, -0.04178556799888611, ...
8c83de1e8186254b154b40cec4beb82477aea3ab
subsection
13
33
Finding efficiency
Google Scholar is a excellent resource for finding articles that suit your research tastes or articles that relate to other topics through citations. This is why scraping it seemed ideal for my data-set needs as a reliable product. Although upon further investigation citation counts were susceptible to manipulation by ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1145/1810617.1810683", "end": 346, "openalex_id": "https://openalex.org/W2013638680", "raw": "Joeran Beel and Bela Gipp. On the robustness of google scholar against spam. In Proceedings of the 21st ACM Conference on Hypertext and Hyperme...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.058183372020721436, -0.028511684387922287, -0.007822397165000439, -0.010707144625484943, -0.00221507390961051, -0.02773326076567173, 0.0011313857976347208, -0.01265320461243391, -0.03107590414583683, 0.03293801471590996, 0.01549216266721487, -0.0016875391593202949, -0.008875559084117413, ...
08ea85d119a15a41bb79f0db44aa097ceb65063f
subsection
14
33
DBLP
On account of most of the scrapers having id's for each of their BibTeX entries, I was on average getting 1 BibTeX entry per 1 request which is good depending on the speed of your requests and if you can cut the time down by running things in parallel or concurrently. DBLP, which is a computer science bibliography libr...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.03125462681055069, -0.016832934692502022, -0.07575584203004837, 0.006218877155333757, -0.008416467346251011, -0.004326506983488798, 0.0102401627227664, -0.01637510396540165, 0.03360482677817345, 0.006745383143424988, -0.016985546797513962, -0.0029530127067118883, -0.038244184106588364, ...
7139e39dd905fd1be54aee3e13b66a123e98b55d
subsection
15
33
Scraping under the radar
Automated scraping using your own 'bot' violates almost every single search engine's terms of service especially if your bot doesn't even look at their robot.txt file which outlines the terms and rules that bots must follow when using the site. "The search engines aren't naive. Google knows every search engine operator...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 409, "openalex_id": "", "raw": "How to: Scrape search engines without pissing them off, Feb 2017. URL https://searchnewscentral.com/blog/2011/09/28/how-to-scrape-se arch-engines-without-pissing-them-off/.", "source_ref_id": ...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.0013144755503162742, -0.033699411898851395, -0.05201430991292, 0.027792857959866524, -0.0066467816941440105, -0.0168344434350729, 0.0034645681735128164, -0.008646158501505852, -0.0056280153803527355, 0.05012177303433418, -0.024648800492286682, -0.004540568683296442, -0.022512061521410942,...
66121ce9b856e0711ffef1a44e796a48df09de7c
subsection
16
33
Maintenance
On my server I had upwards of five to seven scrapers running at any one time. This meant that I was constantly checking their progress and fixing bugs that would arise quite often. I implemented a simple tracking system so that if any of the scrapers was cut short on execution, it would give information on the error, t...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.030798934400081635, -0.00812707282602787, -0.06471133977174759, 0.028753813356161118, -0.02683078683912754, -0.0356217622756958, -0.018222957849502563, 0.010591903701424599, 0.0026880388613790274, -0.0022664230782538652, -0.007921034470200539, 0.002726194215938449, -0.014102187938988209, ...
055e96482beae3a27492f0d07f1d603e1c79c47c
subsection
17
33
Parsing & Removal of Impurities
BibTeX is a format which is relatively simple to parse in most languages. The issue lies in entries containing special characters which will affect a parsers interpretation of an entry, which in turn will produce errors. There exists many 'Bib cleaners' which attempt to clean BibTeX databases by removing duplicates, ch...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.045640796422958374, -0.02735091932117939, -0.04365774244070053, 0.009686466306447983, 0.003950858023017645, -0.04469503462314606, 0.05152894929051399, -0.013652577996253967, 0.056776419281959534, 0.009450024925172329, -0.027533970773220062, -0.009091549552977085, -0.03444415703415871, -...
fa8f03183a229f269b9637bb83fa974b09e9e658
subsection
18
33
The 'HomePage' problem
Within the DBLP scraper I noticed a large number of similar disparities which had accumulated due to the @misc field type for BibTeX entries.All BibTeX field types can be seen in the B appendix. Misc in particular is the shortening of miscellaneous and is used when the type of the entry is not apparent. Within DBLP the...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.049267567694187164, -0.026007413864135742, -0.05750935152173042, 0.02271069958806038, 0.012790030799806118, -0.01467495784163475, 0.05326635763049126, -0.012454254552721977, 0.03672173619270325, 0.02327541448175907, -0.02672475576400757, -0.037423815578222275, -0.019917650148272514, -0....
a7a1c84260a2c2653df5f991ec0cd7852ccd2a90
subsection
19
33
Citation Styles
When referencing research papers and other scholarly articles, particular styles are used depending on the preference of the university or department. This is why citation parsing for meta-data generation is such a complex issue, of course if the parser knew which citation style is being used it could reverse engineer ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 513, "openalex_id": "", "raw": "Citation style language. URL http://citationstyles.org/.", "source_ref_id": "b57d1ea732ad8f91adc5db88f2218b46e0694598", "start": 480 } ] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.0699906200170517, -0.01704000122845173, -0.05818313732743263, 0.04893851652741432, -0.002492309780791402, -0.035483475774526596, 0.024591298773884773, -0.007566553540527821, 0.03743613511323929, 0.009099695831537247, -0.006765658501535654, 0.02437772788107395, -0.00013205237337388098, 0...
092352c882cf225e4c9fb3294773173536ea75db
subsection
20
33
Annotating Citation References
Annotated Citation References are references which contain meta-data fields in XML format which outline the fields that belong to the associated reference tokens. In order to create these annotated references with the correct corresponding fields or labels, the CSL files need to be altered in order to make the addition...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.10944710671901703, -0.014848303981125355, -0.041080813854932785, 0.046330370008945465, -0.004032542929053307, -0.00824057962745428, 0.060003627091646194, -0.0003588562540244311, 0.017869848757982254, -0.004955792799592018, 0.019304320216178894, 0.002399687422439456, -0.0358312614262104, ...
27b0489096ad0b335c4a199ea538c1e2486ee216
subsection
21
33
CiteProc Software & Efficiency Challenges
CiteProc-java is a Citation Style Language (CSL) processor for Java, it interprets and translates CSL styles and generates citations and bibliographies . Citeproc uses a CSL processor in order to make the conversion to a citation. Initializing the processor takes the style as a parameter, which means that you cannot us...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 153, "openalex_id": "", "raw": "Michel Krämer. citeproc-java: A citation style language (csl) processor for java. URL http://michel-kraemer.github.io/citeproc-java/.", "source_ref_id": "0cffb25ebf229bc6fc2a4719f333cbca3b237d...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.058659277856349945, 0.007275185082107782, -0.024812690913677216, 0.060551512986421585, -0.0006318576051853597, -0.0005765403038822114, 0.029665358364582062, 0.014992908574640751, 0.02659810706973076, 0.003696724772453308, 0.016038214787840843, -0.03833301365375519, -0.0410492867231369, ...
dd27252c13881b96378b44fdd2e6598607ad3d3a
subsection
22
33
Structure of the Data-set
The structure of the data-set is dictated by the fields and information needed in order to train and evaluate existing citation tools which are lacking in regards to the extensiveness and diversity of its training data. The BibTeX information at the top of each entry is what is used to create the reference strings whic...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.061326462775468826, -0.03780272975564003, -0.054583605378866196, 0.02463737316429615, 0.002662056591361761, -0.03938928619027138, 0.0685269832611084, 0.01681138388812542, 0.04048766940832138, -0.008657405152916908, -0.028451208025217056, -0.03167008236050606, -0.06584204733371735, 0.011...
206f90ef410a2567af60b3603410740dc12074ae
subsection
23
33
Meta-Data Field consistency & Noise
The need for clean, consistent and clear meta-data fields are necessary when making string comparisons between both the meta-data fields themselves and the inner values of the fields. This means that inaccuracies of meta-data fields can impact assessment percentages of parsing tools. A simple example of this is if the ...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.07263277471065521, -0.018142933025956154, -0.042175836861133575, 0.025284139439463615, -0.00017762412608135492, -0.03540084511041641, 0.034668415784835815, 0.005970078054815531, 0.01305406168103218, 0.00830088835209608, -0.029327770695090294, -0.022018715739250183, -0.029083626344799995, ...
4e0149f102b42950eec179cf145002223eb2f87f
subsection
24
33
Results
The overall scraping resulted in 7-8 text files containing an estimated 2.5 million BibTeX entries in total. Which when multiplied by the number of styles I have prepared for annotation (1600) gives 4 billion possible instances over 2.5 million unique reference strings Currently the data-set contains approximately 20 m...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.03308801352977753, -0.039894863963127136, -0.05311175435781479, 0.046121757477521896, -0.0007206524605862796, -0.04218416288495064, 0.0016616500215604901, -0.017642870545387268, 0.037971850484609604, 0.02336611971259117, -0.026739021763205528, 0.028723081573843956, 0.0071693239733576775, ...
96d12273fdfe3ff2078523f390dd203356260b18
subsection
25
33
Future Work & Data-Set Usability
This Data-set will be used for training citation parsing tools which feature machine learning based approaches. The tools will learn custom parsing rules from this training data. To mention a few of the open sources tools, CERMINE , GROBID , ParsCit .
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.1007/s10032-015-0249-8", "end": 252, "openalex_id": "https://openalex.org/W791527587", "raw": "Dominika Tkaczyk, Pawel Szostek, Mateusz Fedoryszak, Piotr Jan Dendek, and Lukasz Bolikowski. CERMINE: automatic extraction of structured meta...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.0631665512919426, -0.00820097140967846, -0.06432612985372543, 0.05517156049609184, 0.0020712220575660467, -0.048885419964790344, 0.02241344563663006, 0.02628888189792633, 0.02941669337451458, -0.00885704904794693, -0.04659677669405937, -0.014029382728040218, -0.024457966908812523, 0.017...
fe4266b6e522f397e7ea90e107778450a334ec32
subsection
26
33
The Process of Evaluation Testing
'Out of the box' versions of the tools contain pre-trained machine learning models which are used if no other model is provided. The comprehensive data-set is split into 66% training data and 33% evaluation sets. A lot of the tools ask for the annotated references to be formatted in a particular way before being proces...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.054949600249528885, -0.02735275775194168, -0.061417847871780396, 0.027642609551548958, 0.012158476747572422, -0.028130779042840004, 0.03093775361776352, 0.04588795080780983, 0.009282851591706276, 0.006250860169529915, -0.0403045117855072, -0.0073339869268238544, -0.010083755478262901, 0...
96b9babab11b6e6b84cfbac5c853bfbc34fd6288
subsection
27
33
Comparison Method
For each tool, the given output fields that were extracted for each reference are compared against the annotated values of the same reference stored within the data-set which are referred to as the 'ground truth' values. The output fields are altered slightly in order to clean them, which includes cleaning of special c...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 1135, "openalex_id": "", "raw": "Dominika Tkaczyk, Pawel Szostek, Piotr Jan Dendek, Mateusz Fedoryszak, and Lukasz Bolikowski. Cermine–automatic extraction of metadata and references from scientific literature. In Document Analysi...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.05107647553086281, -0.01902400515973568, -0.04067200794816017, 0.01665935292840004, -0.004942884668707848, -0.0957455113530159, 0.0609317347407341, 0.020564841106534004, 0.003644233336672187, 0.036247823387384415, -0.039482057094573975, 0.03817005455493927, -0.00919925794005394, 0.01858...
cd3ab12b2928e097d4f9c2839b5ce55eecdd5aa1
subsection
28
33
Outlook
The scraping was cut short because of time constraints on the project. Additionally the scarcity of available disk and memory on my server was a contributing factor. My server was running out of disk due to the scale of these entries and the amount of data that was being stored in text files from the BibTeX entries. I ...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ 0.008827010169625282, -0.012336453422904015, -0.016433345153927803, 0.024398254230618477, 0.0010766740888357162, -0.032927725464105606, 0.0038107966538518667, -0.014350567944347858, 0.02021743915975094, 0.006091934163123369, -0.0013608626322820783, -0.0270227063447237, -0.004035858903080225,...
cea9a295db31e0fc0f0d1304c0cf9b57a8cd1666
subsection
29
33
Conclusion
Through research of countless scientific papers in this area and maneuvering through the struggles of scraping, I can see why open source data sets for citation parsing tools are sparse. Regardless of this, I am positive that this data set will continue to grow and increase performance for open source meta data extract...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.1802.01168", "end": 464, "openalex_id": "https://openalex.org/W4297693864", "raw": "Dominika Tkaczyk, Andrew Collins, Paraic Sheridan, and Joeran Beel. Machine learning vs. rules and out-of-the-box vs. retrained: An evaluatio...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.07828311622142792, 0.0029592602513730526, -0.06208345293998718, 0.03368065506219864, 0.01214211992919445, -0.03746362775564194, -0.00665833568200469, 0.029165493324398994, 0.028250258415937424, 0.01781657710671425, -0.02526049129664898, -0.015818314626812935, -0.03892800211906433, 0.001...
4e69136111e72ff25fa23dde5ba57084a5e9c040
subsection
30
33
Terms
1.5BibTeX - Bibliography Type SettingDBLP - The dblp computer science bibliography is the on-line reference for open bibliographic information on computer science journals and proceedingsCCS - The Collection of Computer Science BibliographiesPubMed - PubMed is a free search engine accessing primarily the MEDLINE databa...
{ "cite_spans": [] }
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.025612521916627884, -0.022697152569890022, -0.04911865293979645, 0.03428231552243233, 0.0029783332720398903, -0.03495391830801964, 0.03965514525771141, -0.0202396921813488, 0.03788455203175545, 0.024986710399389267, -0.015202665701508522, -0.018255408853292465, -0.004502797499299049, -0...
316eb50f12b64127d5dbb7666e041519574c39ea
subsection
31
33
Information on BibTeX
These are the standard entry types of BibTeX entries and the details of what is required or optional for eacharticle An article from a journal or magazine. Required fields: author, title, journal, year. Optional fields: volume, number, pages, month, note. book A book with an explicit publisher. Required fields: author...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 2482, "openalex_id": "", "raw": "Bibtex entry types, field types and usage hints. URL https://www.openoffice.org/bibliographic/bibtex-defs.html.", "source_ref_id": "b14e0ddf0e76b5c6264e632fa477e9c7a0973503", "start": 2...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.050694793462753296, -0.009560592472553253, -0.03970736265182495, 0.0384865365922451, 0.021135548129677773, -0.005451750475913286, 0.03683841973543167, -0.0168931782245636, 0.027651706710457802, -0.0007029286352917552, -0.04602513462305069, 0.002876570913940668, -0.020067324861884117, -0...
eacc71c4f01210c7955e038118ad2b3aaf73c65e
subsection
32
33
Information on BibTeX
If there is also an author field, then the editor field gives the editor of the book or collection in which the reference appears. howpublished How something strange has been published. The first word should be capitalized. institution The sponsoring institution of a technical report. journal A journal name. Abbrevi...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 2485, "openalex_id": "", "raw": "Bibtex entry types, field types and usage hints. URL https://www.openoffice.org/bibliographic/bibtex-defs.html.", "source_ref_id": "b14e0ddf0e76b5c6264e632fa477e9c7a0973503", "start": 2...
1805.04798
Citation Data-set for Machine Learning Citation Styles and Entity Extraction from Citation Strings
[ "Niall Martin Ryan" ]
[ "cs.DL" ]
2,018
en
Computer Science
[ -0.048132482916116714, -0.013643132522702217, -0.02901073358952999, 0.006398079916834831, 0.025424450635910034, -0.020769914612174034, -0.004059365950524807, -0.0013753772946074605, -0.02087673917412758, 0.013093744404613972, -0.027026832103729248, 0.008446838706731796, -0.04263860359787941,...
7e08357c852ec49d57320e6feb119bd48e32a7a8
abstract
0
118
Abstract
This is an exciting time for the study of r-process nucleosynthesis. Recently, a neutron star merger GW170817 was observed in extraordinary detail with gravitational waves and electromagnetic radiation from radio to gamma rays. The very red color of the associated kilonova suggests that neutron star mergers are an impo...
{ "cite_spans": [] }
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.03610030189156532, 0.023527754470705986, -0.027891527861356735, 0.030042897909879684, -0.01585300639271736, 0.02284114621579647, -0.026014799252152443, 0.031049922108650208, 0.022185055539011955, 0.037076812237501144, -0.008422386832535267, -0.02937154844403267, 0.00258622202090919, 0.0...
7a0e21b706fd18e53e4b36eccdbd774aaf3ef9c4
subsection
1
118
Introduction
How were the elements from Iron to Uranium made? The influential National Academy of Science report “Connecting Quarks to the Cosmos” identified this question as one of eleven questions at the intersections of astronomy and physics that are of deep interest and are ripe for answering . Ever since the pioneering works o...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 286, "openalex_id": "https://openalex.org/W3032091194", "raw": "Council N R 2003 Connecting Quarks with the Cosmos: Eleven Science Questions for the New Century (Washington, DC: The National Academies Press) ISBN 978-0-309-07406-3...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ 0.028185300529003143, 0.037875428795814514, -0.024675490334630013, 0.036685142666101456, -0.004551307763904333, 0.002838368061929941, -0.012490345165133476, 0.02980286441743374, 0.021409841254353523, 0.041965119540691376, -0.0072637805715203285, -0.006474073510617018, -0.030764248222112656, ...
54600cea357f86f14ba8c7f9b3ce3cba051b78d1
subsection
2
118
Introduction
Galactic chemical evolution simulations are reviewed in Sec. , while Sec. reviews astrophysical simulations of r-process sites. Given conditions present in a site, one can perform detailed nuclear reaction network simulations to predict nucleosynthetic yields. Here, an accurate understanding of the relevant nuclear phy...
{ "cite_spans": [] }
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.04033594951033592, 0.027292365208268166, -0.012318940833210945, 0.030358750373125076, -0.0020404437091201544, -0.04793326556682587, 0.005408129189163446, -0.0030568516813218594, 0.010419611819088459, 0.021129081025719643, -0.002358905505388975, -0.006731557194143534, -0.0424107164144516, ...
251b1fd6a19558c59144f85666a7103491a0c4e8
subsection
3
118
Observations of stellar abundances
The heavy elements in the atmospheres of most late-type (F-G-K) stars, which have effective temperatures of \approx  4000–7000 K, reflect the stars' natal compositions, and are untouched by the products of nuclear burning in the interior. Each star thus retains a chemical memory of the content of one piece of the inter...
{ "cite_spans": [] }
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ 0.007993987761437893, -0.0028585370164364576, -0.04418050870299339, 0.0191611610352993, -0.02178514190018177, -0.04259391874074936, 0.00974839273840189, 0.014279335737228394, -0.025553299114108086, 0.02788742072880268, -0.015126027166843414, 0.0013463156064972281, -0.009786532260477543, 0....
eebd5bd6e09398ef7256237392cc991271da583e
subsection
4
118
Measuring detailed
Late-type stars are the only sites beyond the Solar System where detailed chemical abundance patterns for large numbers of elements can be derived. When metal-poor stars that are highly enhanced in r-process elements are identified, detailed abundance analyses based on high-resolution spectra, model atmospheres, and at...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 562, "openalex_id": "", "raw": "Barklem P S, Christlieb N, Beers T C, Hill V, Bessell M S, Holmberg J, Marsteller B, Rossi S, Zickgraf F J and Reimers D 2005 Astron. Astrophys. 439 129–151", "source_ref_id": "cfad90f615c371b...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ 0.007808192167431116, 0.027738725766539574, -0.05013721436262131, 0.035489700734615326, -0.0035760528407990932, -0.02795233577489853, 0.00450487295165658, 0.057247359305620193, -0.02610613778233528, 0.02955440804362297, -0.016356388106942177, -0.030805548653006554, -0.02189497835934162, 0....
0ad7f3f6a4d5e5aa05da640333e0116a8d5cb7f4
subsection
5
118
Measuring detailed
The relative abundances and locations (in mass) of the r-process peaks are sensitive to the conditions (e.g., ), so they are especially valuable probes of the r-process. Many of the radioactive progenitor nuclei for elements at the first and second r-process peak can be produced by current radioactive beam facilities i...
{ "cite_spans": [ { "arxiv_id": "", "doi": "10.48550/arxiv.0709.1535", "end": 169, "openalex_id": "https://openalex.org/W1927953340", "raw": "Kratz K L, Bitouzet J P, Thielemann F K, Moeller P and Pfeiffer B 1993 Astrophys. J. 403 216–238", "source_ref_id": "d2cc64cd00aa2cbfb41...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ 0.004711291287094355, -0.01956234872341156, -0.03912469744682312, 0.032227519899606705, -0.01553390920162201, -0.00749228848144412, 0.004291662015020847, 0.01782279647886753, 0.02362130768597126, 0.023987529799342155, -0.04812765121459961, -0.02026427537202835, -0.010429692454636097, 0.035...
f62f488c69db0240285a1477a1d643ebe9586706
subsection
6
118
Deviations from the
One key observational result is that the r-process pattern is robust from one star to another, and agrees well with the Solar System r-process residuals, for elements at and between the second and third r-process peaks (Ba to Au, e.g., , ). Sometimes this so-called “universality” extends to lighter r-process elements a...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 240, "openalex_id": "https://openalex.org/W3030112318", "raw": "Roederer I U, Lawler J E, Cowan J J, Beers T C, Frebel A, Ivans I I, Schatz H, Sobeck J S and Sneden C 2012 Astrophys. J. Lett. 747 L8", "source_ref_id": "87ddb...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ 0.018358908593654633, -0.0030044973827898502, -0.03818286955356598, 0.024905851110816002, 0.005566427484154701, -0.03146805614233017, -0.02750021032989025, 0.0670870840549469, -0.029270479455590248, 0.030323484912514687, -0.01927456445991993, -0.03854912891983986, -0.03922061249613762, 0.0...
e1952faf3fe2698612c55061e6bdba5a41e06f86
subsection
7
118
Environmental constraints on the
Stars like HD 108317 or CS 22892-052 are located in the field, unaffiliated with any known stellar cluster, stream, or galaxy. This limits their utility in terms of constraining the site of the r-process based on its environment. Recently, the lowest-luminosity galaxies known—also called ultra-faint dwarf galaxies, or ...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 493, "openalex_id": "", "raw": "Bovill M S and Ricotti M 2011 Astrophys. J. 741 17", "source_ref_id": "ab6477381c24d1a92229f1c5776e0c1bad77fa82", "start": 230 }, { "arxiv_id": "", "doi": "10.1146/an...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.04975570738315582, 0.03205122426152229, -0.025412041693925858, 0.028159288689494133, -0.014079644344747066, 0.018253933638334274, -0.04276548698544502, 0.07875443249940872, 0.01494197454303503, 0.010057979263365269, -0.039743516594171524, -0.038339368999004364, -0.01259155198931694, 0.0...
59b7ed33a6f3c1e9eb945dc01c07c5d5d5150053
subsection
8
118
Evidence from elements not produced by the
Another observational approach to identify the astrophysical site(s) of the r-process is to consider the light elements that could be produced along with the r-process. Are the abundance ratios among elements from C to Zn (6 \le Z \le  30) statistically different in stars with high levels of r-process enhancement and t...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 457, "openalex_id": "", "raw": "Roederer I U, Cowan J J, Preston G W, Shectman S A, Sneden C and Thompson I B 2014 Mon. Not. R. Astron. Soc. 445 2970–2984", "source_ref_id": "56be5d188b13a15a89a28fdd52215a74d8d7722c", ...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.05969274789094925, 0.00412371288985014, -0.0204621609300375, 0.0195924062281847, -0.0037613145541399717, 0.012764055281877518, -0.029937928542494774, 0.04644804075360298, -0.0025196229107677937, 0.006820720620453358, -0.017517197877168655, -0.02764909528195858, -0.029998963698744774, 0....
99891db330be3fa4be0613936c448954986b8e5c
subsection
9
118
Stars with low levels of
Highly r-process enhanced stars, as highlighted in the previous sections, comprise only a few percent of the local Galactic halo field population, and only a small fraction of UFD galaxies boast large numbers of highly r-process enhanced stars. In these stars, elements produced by the r-process are still in the minorit...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 641, "openalex_id": "", "raw": "Roederer I U 2013 Astron. J. 145 26", "source_ref_id": "ef2de594a35c14267dbb2051991baab880808cf7", "start": 484 }, { "arxiv_id": "", "doi": "10.48550/arxiv.0709.1535"...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.025425845757126808, 0.010950206778943539, -0.03586478903889656, 0.04489966109395027, 0.01086626760661602, -0.022617708891630173, -0.008729642257094383, 0.06330516189336777, 0.001568130566738546, 0.03302612900733948, -0.019275415688753128, -0.03760461136698723, -0.02016058936715126, 0.04...
09f2b04d09b3165dd0c284ce13cb30077416aaf3
subsection
10
118
Multi-messenger observations of possible
Observations of energetic astronomical events with not just photons, but also neutrinos or gravitational waves, can provide especially important information on the r-process. This is because neutrinos and gravitational waves come from deep within an astronomical event and may directly probe the extreme conditions that ...
{ "cite_spans": [] }
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.010416500270366669, 0.0022626316640526056, -0.011973251588642597, 0.05677564814686775, -0.0018343342235311866, -0.05689774826169014, -0.019703593105077744, 0.01686481013894081, 0.008600289933383465, 0.038460925221443176, -0.022542374208569527, -0.0302498210221529, -0.043131180107593536, ...
cd10ebdb3edbbc963f4ed262ede2872164f4a8ba
subsection
11
118
Neutrinos from core collapse supernovae
One frequently studied theoretical r-process site is the neutrino-driven wind during a core collapse supernova (CCSN). Here intense neutrino and antineutrino fluxes blow baryons off of the protoneutron star and determine the ratio of neutrons to protons in the wind. Antineutrinos capture on protons to make neutrons\bar...
{ "cite_spans": [ { "arxiv_id": "", "doi": "", "end": 2356, "openalex_id": "", "raw": "McLaughlin G C, Fetter J M, Balantekin A B and Fuller G M 1999 Phys. Rev. C 59 2873–2887", "source_ref_id": "576c02e6f005b95526831a648f1a31b176ad9572", "start": 2268 }, { ...
10.1088/1361-6471/ab0849
1805.04637
r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos
[ "C. J. Horowitz", "A. Arcones", "B. Côté", "I. Dillmann", "W. Nazarewicz", "I. U. Roederer", "H. Schatz", "A. Aprahamian", "D. Atanasov", "A. Bauswein", "J. Bliss", "M. Brodeur", "J. A. Clark", "A. Frebel", "F. Foucart", "C. J. Hansen", "O. Just", "A. Kankainen", "G. C. McLaughli...
[ "astro-ph.SR", "nucl-ex", "nucl-th" ]
2,018
en
Physics
[ -0.04217373952269554, 0.01177172176539898, -0.0296772513538599, 0.021254710853099823, -0.015578649938106537, -0.04134979471564293, -0.03396481275558472, 0.065793476998806, 0.029112696647644043, 0.042509421706199646, -0.004276118241250515, -0.021315742284059525, -0.02358921803534031, 0.0295...