betaFeedback

WorksComputer Science

Parallel Broyden Methods for Efficiently Evaluating Nonlinear State Space Models

Abstract

Nonlinear state space models are typically evaluated sequentially, limiting their ability to exploit modern parallel hardware. Recent work has shown that state space models can be evaluated in parallel by reformulating evaluation as a root-finding problem, which can be solved with a parallel form of Newton's method. However, exact Newton methods require expensive multiplication of dense Jacobian matrices, and quasi-Newton methods based on diagonal approximations, while more scalable, are often slow to converge because they neglect interactions across state dimensions. We introduce a novel quasi-Newton approach based on Broyden's method, which captures coupling terms with diagonal-plus-low-rank approximations to blocks of the Jacobian. We construct the approximations from trajectory secants, avoiding automatic differentiation while retaining the favorable scaling of diagonal approximations and remaining compatible with parallel scan through rank compression. We prove finite-step recovery of the exact trajectory under the compressed scan, and show that convergence is tracked by the compression error and by the block approximation error along the current trajectory error, rather than in spectral norm. Empirically, we show that our method is advantageous when interactions are approximately low rank or when Jacobian evaluation is expensive: on low-rank RNNs, parallel MCMC algorithms, and denoising diffusion models, it converges in fewer iterations than other quasi-Newton methods and is the fastest parallel method when Jacobian evaluation is expensive or state dimensions are large.

Introduction

Parallel computation has been a key driver of recent advances in machine learning. Architectures such as transformers [1] and linear state space models [2, 3, 4, 5] are explicitly designed to exploit parallelism over sequence length. In contrast, general nonlinear state space models, including recurrent neural networks [6, 7, 8], diffusion samplers [9, 10, 11, 12], and Markov chain Monte Carlo methods [13, 14, 15, 16, 17], seem inherently sequential, requiring time that scales linearly with trajectory length. This sequential dependency limits their ability to exploit modern parallel hardware.

Recent work, including DEER [18] and DeepPCR [19], has addressed this challenge by reformulating trajectory computation as a root-finding problem over the full sequence. In this formulation, the residual Jacobian has a block-bidiagonal structure, so Newton updates reduce to solving a linear recurrence that can be evaluated with logarithmic parallel depth via parallel scan [20]. This makes it possible to replace sequential evaluation with iterative parallel trajectory refinement.

However, exact Newton updates face two scalability bottlenecks. First, they require storing and multiplying dense dynamics Jacobian matrices, leading to memory and computational costs that scale poorly with state dimension. Second, evaluating these Jacobians with automatic differentiation can be expensive when the transition map is a large neural network or simulator. Scalable quasi-Newton variants address the first bottleneck by replacing each block with a diagonal approximation, reducing memory and computation to scale linearly with state dimension [21]. However, these methods still require estimating or computing the diagonal of the Jacobian, and therefore do not fully remove the cost of Jacobian evaluation. Moreover, the diagonal approximation discards dependencies across state dimensions, which can hinder convergence when the dynamics exhibit strong coupling [22].

In this work, we ask: Can we construct richer Jacobian approximations that retain the scalability of diagonal quasi-Newton while capturing structured cross-dimensional coupling and avoiding expensive Jacobian evaluation? We answer this question by introducing Parallel Broyden methods, a family of quasi-Newton methods based on Broyden’s method that can be parallelized over the sequence length. Broyden’s method is a classical quasi-Newton approach for updating Jacobian approximations from secant information [23], without automatic differentiation. Broyden-style updates underlie several widely used optimization methods, including BFGS and L-BFGS [24, 25]. Our approach targets regimes where cross-dimensional interactions are important but can be captured by low-rank corrections, or where automatic differentiation through the transition map is costly.

Our contributions are as follows:

  • We introduce Parallel Broyden, a family of quasi-Newton methods for parallel nonlinear trajectory evaluation that build scan-compatible diagonal-plus-low-rank Jacobian approximations from trajectory secants, avoiding costly automatic differentiation.

  • We develop a compressed approximate scan that controls the rank of diagonal-plus-low-rank blocks while preserving logarithmic parallel depth and scaling linearly with state dimension for small rank. We prove finite-step recovery of the exact trajectory even with compression.

  • We extend the convergence analysis of parallel quasi-Newton methods to compressed scans, and show theoretically and empirically that convergence is tracked by the block approximation error along the current trajectory error, rather than in spectral norm.

  • We show empirically on low-rank RNNs and hierarchical Gibbs sampling that our approach captures low-rank coupling more effectively than diagonal quasi-Newton, and on diffusion models that it avoids costly differentiation through the denoiser, reducing iteration counts relative to diagonal quasi-Newton and improving runtime over exact and diagonal methods when Jacobian computation or scaling costs dominate, with speedups of up to 30×30\times over sequential evaluation. Code is available at https://github.com/ianctanoh/BroyDEER.

Problem Statement and Related Work

Parallel computation over sequences is central to modern sequence modeling. Structured state space models such as S4, S5, and Mamba exploit algebraic structure for efficient long-sequence computation [2, 4, 3]. For general nonlinear recurrences, recent work instead recasts trajectory evaluation as a nonlinear system solved by iterative refinement, including DEER and DeepPCR [18, 19], Picard iteration for parallel diffusion sampling [26, 27], parallel-in-time methods for ODE solvers [28], and MCMC [29]. Recent unifying work views these approaches as quasi-Newton iterations on the full trajectory residual [22]. We follow this perspective, using Broyden-style secant updates to obtain scalable quasi-Newton approximations for parallel trajectory evaluation.

Following the notation of Gonzalez et al. [22], we consider a DD-dimensional nonlinear state-space model with time-indexed transition maps ft:RD→RDf_t : \mathbb{R}^{D} \to\mathbb{R}^{D}, where the state at time tt is xt∈RDx_t \in\mathbb{R}^{D}. The maps ftf_t may represent, for example, the transition of a recurrent neural network, one step of a numerical simulator, a sampling update, or a diffusion model denoising step. Given an initial state x0x_0, our goal is to compute the trajectory x1:T∈RT×Dx_{1:T} \in\mathbb{R}^{T \times D} satisfying

xt=ft(xt−1),t=1,…,T.(1)x_t = f_t(x_{t-1}), \quad t = 1,\ldots,T. \tag*{(1)}

Although this recurrence is naturally evaluated sequentially, our aim is to expose structure that permits parallel computation across the time axis. For clarity, we absorb input dependence into the time-indexed transition maps ftf_t.

Residual formulation. Rather than solving (1) sequentially, we recast trajectory computation as a root-finding problem. For any candidate trajectory x1:Tx_{1:T}, define the block residual

rt(x1:T):=xt−ft(xt−1),t=1,…,T.(2)r_t(x_{1:T}) := x_t - f_t(x_{t-1}), \quad t = 1,\ldots,T. \tag*{(2)}

and stack these blocks into r(x1:T)∈RT×D\mathbf{r}(\mathbf{x}_{1:T}) \in\mathbb{R}^{T \times D}. The true trajectory x1:T⋆\mathbf{x}^{\star}_{1:T} is characterized by

r(x1:T⋆)=0,(3)\mathbf{r}(\mathbf{x}^{\star}_{1:T}) = 0, \tag*{(3)}

with each residual block measuring the one-step prediction error at time tt. Our goal is to solve (3) in sublinear parallel depth while keeping the per-iteration cost favorable in the state dimension.

Exact parallel Newton [18, 19]. The root-finding formulation (3) allows trajectory evaluation to be performed by Newton iteration over the full sequence. Starting from a candidate sequence x1:T(i)\mathbf{x}^{(i)}_{1:T}, Newton’s method computes a correction Δx(i)\Delta\mathbf{x}^{(i)} that locally reduces all residuals r1,…,rT\mathbf{r}_1,\ldots,\mathbf{r}_T simultaneously by solving

J(x1:T(i))Δx(i)=−r(x1:T(i)),(4)\mathbf{J}\left(\mathbf{x}^{(i)}_{1:T}\right)\Delta\mathbf{x}^{(i)} = -\mathbf{r}\left(\mathbf{x}^{(i)}_{1:T}\right), \tag*{(4)}

where J(x1:T(i))\mathbf{J}(\mathbf{x}^{(i)}_{1:T}) is the Jacobian of the residual map with respect to the trajectory variables. This Jacobian has special structure because each residual block rt=xt−ft(xt−1)\mathbf{r}_t = \mathbf{x}_t - f_t(\mathbf{x}_{t-1}) depends only on the current and previous states. Treating x1:T\mathbf{x}_{1:T} as a flattened vector in RTD\mathbb{R}^{TD} and defining

At(i):=∂ft∂xt−1(xt−1(i))∈RD×D,\mathbf{A}^{(i)}_t := \frac{\partial f_t}{\partial\mathbf{x}_{t-1}}\left(\mathbf{x}^{(i)}_{t-1}\right) \in\mathbb{R}^{D \times D},

the residual Jacobian is block lower bidiagonal:

J(x1:T(i)):=∂r∂x1:T(x1:T(i))=(ID0⋯00−A2(i)ID⋯000−A3(i)⋱00⋮⋮⋱ID000⋯−AT(i)ID).\mathbf{J}\left(\mathbf{x}^{(i)}_{1:T}\right) := \frac{\partial\mathbf{r}}{\partial\mathbf{x}_{1:T}}\left(\mathbf{x}^{(i)}_{1:T}\right) = \begin{pmatrix} \mathbf{I}_D & 0 & \cdots& 0 & 0 \\ -\mathbf{A}^{(i)}_2 & \mathbf{I}_D & \cdots& 0 & 0 \\ 0 & -\mathbf{A}^{(i)}_3 & \ddots& 0 & 0 \\ \vdots& \vdots& \ddots& \mathbf{I}_D & 0 \\ 0 & 0 & \cdots& -\mathbf{A}^{(i)}_T & \mathbf{I}_D \end{pmatrix}.

Solving (4) is therefore equivalent to evaluating a linear recurrence over time, which can be computed with logarithmic parallel depth using parallel scan, a standard primitive for prefix computation.

Diagonal quasi-Newton [21]. Exact Newton is costly in high dimensions because it must form, store, and compose dense dynamics Jacobian blocks At(i)\mathbf{A}^{(i)}_t inside the parallel scan, leading to O(TD2)O(TD^2) memory and O(TD3)O(TD^3) work. Diagonal quasi-Newton reduces this cost by replacing each block with a diagonal approximation. Since diagonal matrices are closed under multiplication, the scan remains exact under the approximation while reducing memory and computation to scale linearly in DD per block. In practice, this approach is most effective when the dynamics Jacobian is well approximated by its diagonal and when the cost of dense Newton-style updates is prohibitive at scale [22].

Beyond the diagonal approximation. However, the diagonal approximation may become too crude when the dynamics are strongly coupled across dimensions. This occurs, for example, in hierarchical Gibbs samplers, where updating one latent variable can change the conditional distribution of many others. Moreover, Gonzalez et al. [22] bound the convergence rate of quasi-Newton updates in terms of the spectral-norm error of the block Jacobian approximation, suggesting that more accurate block approximations lead to faster convergence. This motivates the search for richer quasi-Newton block approximations that better capture these interactions while retaining the O(TD)O(TD) memory and work of diagonal quasi-Newton methods.

Parallel Broyden methods for scalable sequence evaluation

We now introduce Broyden-based quasi-Newton approximations for parallel nonlinear trajectory evaluation. Figure 4 (Appendix § 7) gives an overview of the method. The methods, which we refer to as Parallel Broyden methods, construct diagonal-plus-low-rank (DPLR) approximations of local dynamics Jacobian blocks from trajectory secants rather than automatic differentiation. We first describe the rank-one Broyden updates underlying this construction, then we present a multi-secant variant that we call Accelerated Parallel Broyden. We then show how these DPLR blocks can be made compatible with logarithmic-depth parallel scan through rank compression. Finally, we show that the compressed scan preserves finite-step recovery of the exact trajectory, and analyze how block approximation error and rank compression affect the local convergence rate.

Figure 4. Overview of Parallel Broyden. (a) Evaluating the sequence is cast as solving r(x1:T)=0r(x_{1:T}) = 0 with rt=xt−ft(xt−1)r_t = x_t - f_t(x_{t-1}). Each Newton step solves a block-bidiagonal system J Δx=−r\mathbf{J}\,\Delta x = -r, which is equivalent to the linear recurrence Δxt=AtΔxt−1−rt\Delta x_t = A_t \Delta x_{t-1} - r_t with Jacobians At=∂ft/∂xt−1A_t = \partial f_t/\partial x_{t-1}. (b) Each dense Jacobian AtA_t is replaced by a diagonal-plus-low-rank (DPLR) matrix A~t=Λt+UtVt⊤\widetilde{A}_t = \Lambda_t + U_tV_t^\top, with Λt\Lambda_t diagonal and UtU_t, Vt∈RD×mV_t \in\mathbb{R}^{D \times m}. It is fitted without automatic differentiation from secants pt=xt−1(i+1)−xt−1(i)p_t = x_{t-1}^{(i+1)} - x_{t-1}^{(i)} and qt=ft(xt−1(i+1))−ft(xt−1(i))q_t = f_t(x_{t-1}^{(i+1)}) - f_t(x_{t-1}^{(i)}), either by rank-one Broyden updates (Parallel Broyden) or by fitting the last mm secants jointly (Accelerated Parallel Broyden). (c) The recurrence is solved with a parallel scan over the A~t\widetilde{A}_t. Each merge multiplies two DPLR matrices, which raises the rank of the low-rank part to at most 2m2m, then compresses it back to rank mm with randomized compression, so A~j:i≈A~j⋯A~i\widetilde{A}_{j:i} \approx\widetilde{A}_j \cdots\widetilde{A}_i. Each merge costs O(Dm2)O(Dm^2) instead of O(D3)O(D^3), giving O(log⁡T)O(\log T) depth and O(TDm)O(TDm) memory. The iterate is updated as x(i+1)=x(i)+Δxx^{(i+1)} = x^{(i)} + \Delta x until ∥r∥≤ϵ\lVert r\rVert\le\epsilon.

DPLR Jacobian approximation with Broyden’s method

Parallel Broyden applies Broyden’s method blockwise to the local dynamics functions. Rather than forming Jacobians with automatic differentiation, it updates each block approximation using finite differences between successive trajectory iterates. Since Broyden updates are rank-one corrections, initializing with a diagonal baseline naturally yields diagonal-plus-low-rank approximations without automatic differentiation.

Blockwise Broyden update. After computing the iterate x1:T(i+1)x_{1:T}^{(i+1)}, we form a secant pair for each transition ftf_t:

pt(i):=xt−1(i+1)−xt−1(i),qt(i):=ft(xt−1(i+1))−ft(xt−1(i)).p_t^{(i)} := x_{t-1}^{(i+1)} - x_{t-1}^{(i)}, \qquad q_t^{(i)} := f_t(x_{t-1}^{(i+1)}) - f_t(x_{t-1}^{(i)}).

References

  1. [1]Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017.arxiv.org/abs/1706.03762
  2. [2]Albert Gu, Karan Goel, and Christopher Ré. Efficiently modeling long sequences with structured state spaces. arXiv preprint arXiv:2111.00396, 2021.arxiv.org/abs/2111.00396
  3. [3]Albert Gu. Modeling sequences with structured state spaces. Stanford University, 2023.arxiv.org/abs/2111.00396
  4. [4]Jimmy TH Smith, Andrew Warrington, and Scott W Linderman. Simplified state space layers for sequence modeling. arXiv preprint arXiv:2208.04933, 2022.arxiv.org/abs/2208.04933
  5. [5]Eric Martin and Chris Cundy. Parallelizing linear recurrent neural nets over sequence length. arXiv preprint arXiv:1709.04057, 2017.arxiv.org/abs/1709.04057
  6. [6]Alex Graves. Long short-term memory. Supervised sequence labelling with recurrent neural networks, pages 37–45, 2012.DOI
  7. [7]Kyunghyun Cho, Bart Van Merriënboer, Çağlar Gülçehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. Learning phrase representations using rnn encoder–decoder for statistical machine translation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), pages 1724–1734, 2014.arxiv.org/abs/1406.1078
  8. [8]Ibomoiye Domor Mienye, Theo G Swart, and George Obaido. Recurrent neural networks: A comprehensive review of architectures, variants, and applications. Information, 15(9):517, 2024.
  9. [9]Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2020.arxiv.org/abs/2010.02502
  10. [10]Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020.arxiv.org/abs/2011.13456
  11. [11]Tero Karras, Miika Aittala, Timo Aila, and Samuli Laine. Elucidating the design space of diffusion-based generative models. Advances in neural information processing systems, 35: 25655–26577, 2022.arxiv.org/abs/2206.00364
  12. [12]Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020.arxiv.org/abs/2006.11239
  13. [13]Stuart Geman and Donald Geman. Stochastic relaxation, gibbs distributions, and the bayesian restoration of images. IEEE Transactions on pattern analysis and machine intelligence, (6): 721–741, 1984.
  14. [14]W Keith Hastings. Monte carlo sampling methods using markov chains and their applications. 1970.
  15. [15]Nicholas Metropolis, Arianna W Rosenbluth, Marshall N Rosenbluth, Augusta H Teller, and Edward Teller. Equation of state calculations by fast computing machines. The journal of chemical physics, 21(6):1087–1092, 1953.
  16. [16]Radford M Neal. Mcmc using hamiltonian dynamics. Handbook of markov chain monte carlo, pages 47–95, 2011.DOI
  17. [17]Gareth O Roberts and Jeffrey S Rosenthal. Optimal scaling of discrete approximations to langevin diffusions. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 60(1):255–268, 1998.DOI
  18. [18]Yi Heng Lim, Qi Zhu, Joshua Selfridge, and Muhammad Firmansyah Kasim. Parallelizing non-linear sequential models over the sequence length. arXiv preprint arXiv:2309.12252, 2023.arxiv.org/abs/2309.12252
  19. [19]Federico Danieli, Miguel Sarabia, Xavier Suau Cuadros, Pedro Rodriguez, and Luca Zappella. Deeppcr: Parallelizing sequential operations in neural networks. Advances in Neural Information Processing Systems, 36:47598–47625, 2023.arxiv.org/abs/2309.16318
  20. [20]Guy E Blelloch. Scans as primitive parallel operations. IEEE Transactions on computers, 38 (11):1526–1538, 1989.DOI
  21. [21]Xavier Gonzalez, Andrew Warrington, Jimmy T Smith, and Scott W Linderman. Towards scalable and stable parallelization of nonlinear rnns. Advances in Neural Information Processing Systems, 37:5817–5849, 2024.arxiv.org/abs/2407.19115
  22. [22]Xavier Gonzalez, E Kelly Buchanan, Hyun Dong Lee, Jerry Weihong Liu, Ke Alexander Wang, David M Zoltowski, Leo Kozachkov, Christopher Ré, and Scott W Linderman. A unifying framework for parallelizing sequential models with linear dynamical systems. arXiv preprint arXiv:2509.21716, 2025.arxiv.org/abs/2509.21716
  23. [23]Charles G Broyden. A class of methods for solving nonlinear simultaneous equations. Mathematics of computation, 19(92):577–593, 1965.DOI
  24. [24]Charles George Broyden. The convergence of a class of double-rank minimization algorithms 1. general considerations. IMA Journal of Applied Mathematics, 6(1):76–90, 1970.DOI
  25. [25]Jorge Nocedal. Updating quasi-newton matrices with limited storage. Mathematics of computation, 35(151):773–782, 1980.DOI
  26. [26]Andy Shih, Suneel Belkhale, Stefano Ermon, Dorsa Sadigh, and Nima Anari. Parallel sampling of diffusion models. Advances in Neural Information Processing Systems, 36:4263–4276, 2023.arxiv.org/abs/2305.16317
  27. [27]Zhiwei Tang, Jiasheng Tang, Hao Luo, Fan Wang, and Tsung-Hui Chang. Accelerating parallel sampling of diffusion models. In Forty-first International Conference on Machine Learning, 2024.arxiv.org/abs/2402.09970
  28. [28]Casian Jacob, Hassan Razavi, and Simo Särkkä. A parallel-in-time newton’s method-based ode solver. arXiv preprint arXiv:2511.01465, 2025.
  29. [29]David M Zoltowski, Skyler Wu, Xavier Gonzalez, Leo Kozachkov, and Scott Linderman. Parallelizing MCMC across the sequence length. In The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025.
  30. [30]Haw-ren Fang and Yousef Saad. Two classes of multiescant methods for nonlinear acceleration. Numerical linear algebra with applications, 16(3):197–221, 2009.DOI
  31. [31]Robert B Schnabel. Quasi-newton methods using multiple secant equations. Technical report, 1983.DOI
  32. [32]Rajarshi Saha, Varun Srivastava, and Mert Pilanci. Matrix compression via randomized low rank and low precision factorization. Advances in Neural Information Processing Systems, 36:18828–18872, 2023.arxiv.org/abs/2310.11028
  33. [33]Rafi Witten and Emmanuel Candes. Randomized algorithms for low-rank matrix factorizations: sharp performance bounds. Algorithmica, 72(1):264–281, 2015.arxiv.org/abs/1308.5697
  34. [34]Xavier Gonzalez, Leo Kozachkov, David M Zoltowski, Kenneth L Clarkson, and Scott W Linderman. Predictability enables parallelization of nonlinear state space models. arXiv preprint arXiv:2508.16817, 2025.arxiv.org/abs/2508.16817
  35. [35]Matthijs Pals, A Erdem Sağtekin, Felix Pei, Manuel Gloeckler, and Jakob H Macke. Inferring stochastic low-rank recurrent neural networks from neural data. Advances in Neural Information Processing Systems, 37:18225–18264, 2024.
  36. [36]Friedrich Schuessler, Francesca Mastrogiuseppe, Alexis Dubreuil, Srdjan Ostojic, and Omri Barak. The interplay between randomness and structure during learning in rnns. Advances in neural information processing systems, 33:13352–13362, 2020.arxiv.org/abs/2006.11036
  37. [37]Francesca Mastrogiuseppe and Srdjan Ostojic. Linking connectivity, dynamics, and computations in low-rank recurrent neural networks. Neuron, 99(3):609–623, 2018.DOI
  38. [38]Rainer Engelken, Fred Wolf, and Larry F Abbott. Lyapunov spectra of chaotic recurrent neural networks. Physical Review Research, 5(4):043044, 2023.arxiv.org/abs/2006.02427
  39. [39]Donald B Rubin. Estimation in parallel randomized experiments. Journal of Educational Statistics, 6(4):377–401, 1981.DOI
  40. [40]Andrew Gelman, John B Carlin, Hal S Stern, and Donald B Rubin. Bayesian data analysis. Chapman and Hall/CRC, 1995.
  41. [41]Michael F Hutchinson. A stochastic estimator of the trace of the influence matrix for laplacian smoothing splines. Communications in Statistics-Simulation and Computation, 18(3):1059–1076, 1989.
  42. [42]William Peebles and Saining Xie. Scalable diffusion models with transformers. In Proceedings of the IEEE/CVF international conference on computer vision, pages 4195–4205, 2023.arxiv.org/abs/2212.09748
  43. [43]Carl Eckart and Gale Young. The approximation of one matrix by another of lower rank. Psychometrika, 1(3):211–218, 1936.DOI
  44. [44]Leon Mirsky. Symmetric gauge functions and unitarily invariant norms. The quarterly journal of mathematics, 11(1):50–59, 1960.DOI
  45. [45]W Daniel Hillis and Guy L Steele Jr. Data parallel algorithms. Communications of the ACM, 29(12):1170–1183, 1986.DOI
  46. [46]Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 586–595, 2018.arxiv.org/abs/1801.03924

Paper details

vlnc:W_bm5tuszg
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 7

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Computer Science

Not recorded for this paper

View statement
Looking for related papers…
Contents