pith. sign in

arxiv: 1907.01586 · v1 · pith:MXL67P5Jnew · submitted 2019-07-02 · 💻 cs.CR · cs.LG

Protecting Privacy of Users in Brain-Computer Interface Applications

Anisha Agarwal , Rafael Dowsley , Nicholas D. McKinney , Dongrui Wu , Chin-Teng Lin , Martine De Cock , Anderson C. A. Nascimento This is my paper

Pith reviewed 2026-05-25 10:47 UTC · model grok-4.3

classification 💻 cs.CR cs.LG
keywords privacy-preserving machine learningsecure multiparty computationEEG databrain-computer interfacelinear regressiondrowsiness detectioncryptographic protocols
0
0 comments X

The pith

Secure multiparty computation enables linear regression on EEG signals without revealing individual user data.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper demonstrates cryptographic protocols using secure multiparty computation to perform linear regression over EEG signals from multiple users while ensuring no individual's data is revealed. This approach is crucial because EEG signals contain intimate information that could be misused if exposed. The authors apply it to driver drowsiness estimation, achieving comparable results to standard methods at reasonable computational cost. It represents the first such application to EEG and the largest SMC experiment with 15 players.

Core claim

The authors develop cryptographic protocols based on secure multiparty computation that perform linear regression over EEG signals from many users such that no individual's EEG signals are revealed to anyone else, and apply this to estimate driver drowsiness with performance comparable to the non-private case at reasonable computational cost. This is the first use of commodity-based SMC on EEG data and the largest secret-sharing SMC experiment with 15 players.

What carries the argument

Secure multiparty computation protocols for privacy-preserving linear regression on distributed EEG datasets

If this is right

  • Linear regression models for EEG-based tasks can be trained without exposing raw signals.
  • Drowsiness estimation from EEG can be done privately with similar accuracy to the unencrypted case.
  • The framework supports computations involving up to 15 users with acceptable overhead.
  • Privacy protection is achieved without loss of model utility for the target task.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same approach could apply to other machine learning models beyond linear regression on EEG data.
  • It opens possibilities for cross-institutional collaboration on BCI applications without sharing raw data.
  • Reductions in SMC overhead might enable real-time private BCI inference in the future.

Load-bearing premise

The secure multiparty computation primitives can be efficiently realized for the volume and structure of real EEG datasets while preserving both privacy and the utility of linear regression for the target task.

What would settle it

Running the SMC protocol on real EEG data and finding that the resulting drowsiness prediction model has substantially lower accuracy than a standard linear regression trained on the pooled unencrypted data.

Figures

Figures reproduced from arXiv: 1907.01586 by Anderson C. A. Nascimento, Anisha Agarwal, Chin-Teng Lin, Dongrui Wu, Martine De Cock, Nicholas D. McKinney, Rafael Dowsley.

Figure 1
Figure 1. Figure 1: Training phase of privacy-preserving target-independent LR. [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Inference phase of privacy-preserving target-independent LR. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Training phase of privacy-preserving target-calibrated LR. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Inference phase of privacy-preserving target-calibrated LR. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Different roles in the SMC framework Lynx. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
read the original abstract

Machine learning (ML) is revolutionizing research and industry. Many ML applications rely on the use of large amounts of personal data for training and inference. Among the most intimate exploited data sources is electroencephalogram (EEG) data, a kind of data that is so rich with information that application developers can easily gain knowledge beyond the professed scope from unprotected EEG signals, including passwords, ATM PINs, and other intimate data. The challenge we address is how to engage in meaningful ML with EEG data while protecting the privacy of users. Hence, we propose cryptographic protocols based on Secure Multiparty Computation (SMC) to perform linear regression over EEG signals from many users in a fully privacy-preserving (PP) fashion, i.e.~such that each individual's EEG signals are not revealed to anyone else. To illustrate the potential of our secure framework, we show how it allows estimating the drowsiness of drivers from their EEG signals as would be possible in the unencrypted case, and at a very reasonable computational cost. Our solution is the first application of commodity-based SMC to EEG data, as well as the largest documented experiment of secret sharing based SMC in general, namely with 15 players involved in all the computations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The paper proposes cryptographic protocols based on Secure Multiparty Computation (SMC) to perform linear regression over EEG signals from multiple users in a fully privacy-preserving manner, ensuring individual signals are not revealed. It illustrates the framework by estimating driver drowsiness from EEG data with performance equivalent to the unencrypted case at reasonable computational cost, claiming this as the first commodity-based SMC application to EEG data and the largest secret-sharing SMC experiment with 15 players.

Significance. If the protocols achieve exact utility preservation and scale with reasonable overhead on real EEG volumes, the work would enable collaborative privacy-preserving analysis of sensitive neural data in BCI applications, addressing key privacy risks in ML. The reported scale of the 15-player experiment would be a notable strength if backed by concrete timings and data dimensions.

major comments (3)
  1. [Abstract] Abstract: the claim that SMC-based regression 'allows estimating the drowsiness of drivers from their EEG signals as would be possible in the unencrypted case' lacks any error metrics, accuracy comparisons, or experimental results, which is load-bearing for validating that utility is preserved without approximation.
  2. [Abstract] Abstract: no matrix dimensions (channels, time samples), iteration counts, timing breakdowns, or communication costs are supplied for the 15-player experiment, preventing assessment of whether 'very reasonable computational cost' holds for typical EEG feature matrices (hundreds of channels, thousands of samples across 15+ parties).
  3. [Abstract] Abstract: the manuscript asserts SMC protocols for privacy-preserving linear regression but provides no security proofs, formal privacy analysis, or implementation details, which is essential to substantiate the 'fully privacy-preserving' guarantee.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed review and constructive feedback on our manuscript. We address each major comment point-by-point below, focusing on the abstract as noted. Where appropriate, we indicate willingness to revise for clarity while maintaining the paper's core contributions on SMC for EEG-based linear regression.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that SMC-based regression 'allows estimating the drowsiness of drivers from their EEG signals as would be possible in the unencrypted case' lacks any error metrics, accuracy comparisons, or experimental results, which is load-bearing for validating that utility is preserved without approximation.

    Authors: The abstract is a concise summary; the full manuscript (Section on experimental evaluation) reports that the privacy-preserving regression achieves equivalent performance to the plaintext case on the driver drowsiness EEG dataset, with matching accuracy metrics (e.g., comparable mean squared error or classification rates across the 15 parties). This is not an approximation but exact utility preservation due to the properties of the SMC protocol. We agree the abstract could better support the claim and will revise it to include a brief quantitative reference to the equivalence (e.g., 'with performance metrics matching the unencrypted baseline'). revision: yes

  2. Referee: [Abstract] Abstract: no matrix dimensions (channels, time samples), iteration counts, timing breakdowns, or communication costs are supplied for the 15-player experiment, preventing assessment of whether 'very reasonable computational cost' holds for typical EEG feature matrices (hundreds of channels, thousands of samples across 15+ parties).

    Authors: The manuscript body provides the experimental details for the 15-player setting, including EEG matrix dimensions, iteration counts for the regression, runtime breakdowns, and communication volumes, demonstrating feasibility at reasonable cost for the evaluated scales. The abstract summarizes these findings at a high level. We can partially revise the abstract to include key figures (e.g., approximate matrix sizes and total runtime) to aid assessment without exceeding length limits. revision: partial

  3. Referee: [Abstract] Abstract: the manuscript asserts SMC protocols for privacy-preserving linear regression but provides no security proofs, formal privacy analysis, or implementation details, which is essential to substantiate the 'fully privacy-preserving' guarantee.

    Authors: The protocols rely on established secret-sharing SMC primitives (commodity-based), with privacy following directly from the standard security definitions of the underlying framework (e.g., against semi-honest adversaries). The manuscript emphasizes the novel application to EEG data and the large-scale experiment rather than reproving base SMC results; implementation and protocol details appear in the main text. We do not believe formal proofs are required in the abstract itself, but can add a reference to the security model in a revision if needed. revision: no

Circularity Check

0 steps flagged

No circularity: protocol application is independent of fitted results or self-referential definitions

full rationale

The paper proposes applying existing secure multiparty computation primitives to perform linear regression on EEG data for privacy-preserving drowsiness estimation. No derivation chain reduces a claimed result to its own inputs by construction, no parameters are fitted and then relabeled as predictions, and no load-bearing uniqueness theorems or ansatzes are imported via self-citation. The central contribution is an engineering demonstration whose correctness rests on standard SMC security definitions and empirical timing measurements rather than any self-referential mathematical step.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the applicability and efficiency of existing SMC techniques to EEG data without introducing new cryptographic primitives or parameters.

axioms (1)
  • domain assumption Standard security assumptions of secure multiparty computation (e.g., semi-honest or malicious adversary models) hold for the EEG sharing scenario.
    Privacy guarantees depend on the correctness of the underlying SMC security model.

pith-pipeline@v0.9.0 · 5763 in / 1210 out tokens · 57162 ms · 2026-05-25T10:47:29.925440+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages

  1. [1]

    How susceptible is the brain to the side-channel private information extraction,

    T. Bonaci, J. Herron, and H. Chizeck, “How susceptible is the brain to the side-channel private information extraction,” American Journal of Bioethics, Neuroscience, vol. 6, no. 4, pp. 82–83, 2015

    work page 2015

  2. [2]

    Neural markers of religious conviction,

    M. Inzlicht, I. McGregor, J. B. Hirsh, and K. Nash, “Neural markers of religious conviction,” Psychol. Sci., vol. 20, no. 3, pp. 385–392, 2009

    work page 2009

  3. [3]

    Guessing what’s on your mind,

    M. Poel, C. M ¨uhl, B. Reuderink, and A. Nijholt, “Guessing what’s on your mind,” in Int. Conf. of Brain Informatics , 2010, pp. 180–191

    work page 2010

  4. [4]

    Attentional bias pattern recognition in spiking neural networks from spatio-temporal EEG data,

    Z. G. Doborjeh, M. G. Doborjeh, and N. Kasabov, “Attentional bias pattern recognition in spiking neural networks from spatio-temporal EEG data,” Cognitive Computation, vol. 10, no. 1, pp. 35–48, 2018

    work page 2018

  5. [5]

    Brain- computer-interface-based intervention re-normalizes brain functional network topology in children with attention deficit/hyperactivity disor- der,

    X. Qian, B. R. Y . Loo, F. X. Castellanos, S. Liu, H. L. Koh, X. W. W. Poh, R. Krishnan, D. Fung, M. W. Chee, C. Guan et al. , “Brain- computer-interface-based intervention re-normalizes brain functional network topology in children with attention deficit/hyperactivity disor- der,” Translational Psychiatry, vol. 8, no. 1, p. 149, 2018

    work page 2018

  6. [6]

    Driver drowsiness estimation from EEG signals using online weighted adap- tation regularization for regression (OwARR),

    D. Wu, V . J. Lawhern, S. Gordon, B. J. Lance, and C.-T. Lin, “Driver drowsiness estimation from EEG signals using online weighted adap- tation regularization for regression (OwARR),” IEEE Transactions on Fuzzy Systems, vol. 25, no. 6, pp. 1522–1535, 2017. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING 10

    work page 2017

  7. [7]

    On the feasibility of side-channel attacks with brain-computer interfaces,

    I. Martinovic, D. Davies, M. Frank, D. Perito, T. Ros, and D. Song, “On the feasibility of side-channel attacks with brain-computer interfaces,” in Proc. of the 21st USENIX Security Symposium , 2012

    work page 2012

  8. [8]

    App stores for the brain: Privacy & security in brain-computer interfaces,

    T. Bonaci, R. Calo, and H. J. Chizeck, “App stores for the brain: Privacy & security in brain-computer interfaces,” IEEE Technology and Society Magazine, vol. 34, no. 2, pp. 32–39, 2015

    work page 2015

  9. [9]

    USACM statement on the importance of preserving personal privacy,

    ACM, “USACM statement on the importance of preserving personal privacy,” https://www.acm.org/articles/bulletins/2018/march/ usacm-statement-on-data-privacy, 2018, accessed Mar 30, 2018

    work page 2018

  10. [10]

    Systems and methods for deidentifying entries in a data source,

    L. Sweeney, “Systems and methods for deidentifying entries in a data source,” 2007, US Patent 7,269,578

    work page 2007

  11. [11]

    Cramer, I

    R. Cramer, I. Damg ˚ard, and J. B. Nielsen, Secure Multiparty Computa- tion and Secret Sharing . Cambridge University Press, 2015

    work page 2015

  12. [12]

    Lynx: A framework for privacy preserving machine learning,

    A. Agarwal, K. Saminathan, and S. Bhagat, “Lynx: A framework for privacy preserving machine learning,” https://bitbucket.org/uwtppml

    work page

  13. [13]

    Deep learning with differential privacy,

    M. Abadi, A. Chu, I. Goodfellow, H. B. McMahan, I. Mironov, K. Tal- war, and L. Zhang, “Deep learning with differential privacy,” in 23rd ACM SIGSAC Conf. on Comp. and Comm. Security , 2016, pp. 308–318

    work page 2016

  14. [14]

    Securing data analytics on SGX with randomization,

    S. Chandra, V . Karande, Z. Lin, L. Khan, M. Kantarcioglu, and B. Thu- raisingham, “Securing data analytics on SGX with randomization,” in Eur. Symp. on Research in Comp. Sec. Springer, 2017, pp. 352–369

    work page 2017

  15. [15]

    Practical secure aggregation for privacy-preserving machine learning,

    K. Bonawitz, V . Ivanov, B. Kreuter, A. Marcedone, H. B. McMahan, S. Patel, D. Ramage, A. Segal, and K. Seth, “Practical secure aggregation for privacy-preserving machine learning,” in24th ACM SIGSAC Conf. on Comp. and Comm. Security , 2017, pp. 1175–1191

    work page 2017

  16. [16]

    Privacy-preserving deep learning,

    R. Shokri and V . Shmatikov, “Privacy-preserving deep learning,” in CCS 2015, 2015, pp. 1310–1321

    work page 2015

  17. [17]

    Predicting the brain activation pattern associated with the propositional content of a sentence: Modeling neural representations of events and states,

    J. Wang, V . L. Cherkassky, and M. A. Just, “Predicting the brain activation pattern associated with the propositional content of a sentence: Modeling neural representations of events and states,” Human Brain Mapping, vol. 38, no. 10, pp. 4865–4881, 2017

    work page 2017

  18. [18]

    Sharing deep generative representation for perceived image reconstruction from human brain activity,

    C. Du, C. Du, and H. He, “Sharing deep generative representation for perceived image reconstruction from human brain activity,” in Proc. of International Joint Conf. on Neural Networks , 2017, pp. 1049–1056

    work page 2017

  19. [19]

    Moving magnetoencephalography towards real-world applications with a wearable system,

    E. Boto, N. Holmes, J. Leggett, G. Roberts, V . Shah, S. S. Meyer, L. D. Mu ˜noz, K. J. Mullinger, T. M. Tierney, S. Bestmann et al. , “Moving magnetoencephalography towards real-world applications with a wearable system,” Nature, vol. 555, pp. 657–661, 2018

    work page 2018

  20. [20]

    Highly efficient linear regression outsourcing to a cloud,

    F. Chen, T. Xiang, X. Lei, and J. Chen, “Highly efficient linear regression outsourcing to a cloud,” IEEE Transactions on Cloud Computing, vol. 2, no. 4, pp. 499–508, 2014

    work page 2014

  21. [21]

    Secure multiple linear regres- sion based on homomorphic encryption,

    R. Hall, S. E. Fienberg, and Y . Nardi, “Secure multiple linear regres- sion based on homomorphic encryption,” Journal of Official Statistics , vol. 27, no. 4, pp. 669–691, 2011

    work page 2011

  22. [22]

    Fast and secure linear regression and biometric authentication with security update

    Y . Aono, T. Hayashi, L. T. Phong, and L. Wang, “Fast and secure linear regression and biometric authentication with security update.” IACR Cryptology ePrint Archive , vol. 2015, p. 692, 2015

    work page 2015

  23. [23]

    Privacy-preserving ridge regression on hundreds of millions of records,

    V . Nikolaenko, U. Weinsberg, S. Ioannidis, M. Joye, D. Boneh, and N. Taft, “Privacy-preserving ridge regression on hundreds of millions of records,” in IEEE Symp. on Security and Privacy , 2013, pp. 334–348

    work page 2013

  24. [24]

    Privacy-preserving analysis of vertically partitioned data using secure matrix products,

    A. F. Karr, X. Lin, A. P. Sanil, and J. P. Reiter, “Privacy-preserving analysis of vertically partitioned data using secure matrix products,” Journal of Official Statistics , vol. 25, no. 1, p. 125, 2009

    work page 2009

  25. [25]

    Privacy preserving re- gression modelling via distributed computation,

    A. P. Sanil, A. F. Karr, X. Lin, and J. P. Reiter, “Privacy preserving re- gression modelling via distributed computation,” in 10th ACM SIGKDD Int. Conf. on Knowledge Disc. and Data Mining , 2004, pp. 677–682

    work page 2004

  26. [26]

    Secure regression on distributed databases,

    A. F. Karr, X. Lin, A. P. Sanil, and J. P. Reiter, “Secure regression on distributed databases,” Journal of Computational and Graphical Statistics, vol. 14, no. 2, pp. 263–279, 2005

    work page 2005

  27. [27]

    Privacy-preserving cooperative statistical analysis,

    W. Du and M. J. Atallah, “Privacy-preserving cooperative statistical analysis,” in 17th Annual Comp. Sec. Appl. Conf. , 2001, pp. 102–110

    work page 2001

  28. [28]

    Fast, privacy preserving linear regression over distributed datasets based on pre-distributed data,

    M. De Cock, R. Dowsley, A. C. A. Nascimento, and S. C. Newman, “Fast, privacy preserving linear regression over distributed datasets based on pre-distributed data,” in 8th ACM Workshop on Artificial Intelligence and Security (AISec) , 2015, pp. 3–14

    work page 2015

  29. [29]

    Privacy-preserving distributed linear regression on high- dimensional data,

    A. Gasc ´on, P. Schoppmann, B. Balle, M. Raykova, J. Doerner, S. Zahur, and D. Evans, “Privacy-preserving distributed linear regression on high- dimensional data,” Proceedings on Privacy Enhancing Technologies, vol. 2017, no. 4, pp. 345 – 364, 2017

    work page 2017

  30. [30]

    Privacy-preserving multivariate statistical analysis: Linear regression and classification,

    W. Du, Y . S. Han, and S. Chen, “Privacy-preserving multivariate statistical analysis: Linear regression and classification,” in 4th SIAM International Conference on Data Mining , 2004, pp. 222–233

    work page 2004

  31. [31]

    Tools for privacy preserving distributed data mining,

    C. Clifton, M. Kantarcioglu, J. Vaidya, X. Lin, and M. Y . Zhu, “Tools for privacy preserving distributed data mining,” SIGKDD Explor. Newsl., vol. 4, no. 2, pp. 28–34, 2002

    work page 2002

  32. [32]

    A general survey of privacy- preserving data mining models and algorithms,

    C. C. Aggarwal and S. Y . Philip, “A general survey of privacy- preserving data mining models and algorithms,” in Privacy-Preserving Data Mining. Springer, 2008, pp. 11–52

    work page 2008

  33. [33]

    Practical secure decision tree learning in a teletreatment application,

    S. de Hoogh, B. Schoenmakers, P. Chen, and H. op den Akker, “Practical secure decision tree learning in a teletreatment application,” in International Conference on Financial Cryptography and Data Security. Springer, 2014, pp. 179–194

    work page 2014

  34. [34]

    Privacy-preserving scoring of tree ensembles: A novel framework for AI in healthcare,

    K. Fritchman, K. Saminathan, R. Dowsley, T. Hughes, M. De Cock, A. Nascimento, and A. Teredesai, “Privacy-preserving scoring of tree ensembles: A novel framework for AI in healthcare,” in Proceedings of 2018 IEEE International Conference on Big Data, 2018, pp. 2412–2421

    work page 2018

  35. [35]

    Sharemind: A framework for fast privacy-preserving computations,

    D. Bogdanov, S. Laur, and J. Willemson, “Sharemind: A framework for fast privacy-preserving computations,” in 13th Eur. Symp. on Research in Comp. Sec. , 2008, pp. 192–206

    work page 2008

  36. [36]

    FairplayMP: A system for secure multi-party computation,

    A. Ben-David, N. Nisan, and B. Pinkas, “FairplayMP: A system for secure multi-party computation,” in Proc. of 15th ACM Conference on Computer and Communications Security , 2008, pp. 257–266

    work page 2008

  37. [37]

    Chameleon: A Hybrid Secure Computation Framework for Machine Learning Applications,

    M. Sadegh Riazi, C. Weinert, O. Tkachenko, E. M. Songhori, T. Schnei- der, and F. Koushanfar, “Chameleon: A Hybrid Secure Computation Framework for Machine Learning Applications,” ArXiv e-prints, 2018

    work page 2018

  38. [38]

    Artificial intelligence helps to keep tired drivers awake,

    T. Sandle, “Artificial intelligence helps to keep tired drivers awake,” http://www.digitaljournal.com/tech-and-science/technology/ artificial-intelligence-helps-to-keep-tired-drivers-awake/article/499369

    work page

  39. [39]

    Privacy-preserving linear regression for brain-computer interface applications,

    A. Agarwal, R. Dowsley, N. D. McKinney, D. Wu, C.-T. Lin, M. De Cock, and A. Nascimento, “Privacy-preserving linear regression for brain-computer interface applications,” in Proceedings of 2018 IEEE International Conference on Big Data , 2018, pp. 5260–5261

    work page 2018

  40. [40]

    One-time tables for two-party computation,

    D. Beaver, “One-time tables for two-party computation,” in Computing and Combinatorics. Springer, 1998, pp. 361–370

    work page 1998

  41. [41]

    Efficient unconditionally secure comparison and privacy preserving machine learning classification protocols,

    B. David, R. Dowsley, R. Katti, and A. C. Nascimento, “Efficient unconditionally secure comparison and privacy preserving machine learning classification protocols,” in International Conference on Prov- able Security. Springer, 2015, pp. 354–367

    work page 2015

  42. [42]

    Efficient and private scoring of decision trees, support vector machines and logistic regression models based on pre-computation,

    M. De Cock, R. Dowsley, C. Horst, R. Katti, A. Nascimento, W.- S. Poon, and S. Truex, “Efficient and private scoring of decision trees, support vector machines and logistic regression models based on pre-computation,” IEEE Transactions on Dependable and Secure Computing, vol. 16, no. 2, pp. 217–230, March 2019

    work page 2019

  43. [43]

    Unconditionally secure commitment and oblivious transfer schemes using private channels and a trusted initializer,

    R. L. Rivest, “Unconditionally secure commitment and oblivious transfer schemes using private channels and a trusted initializer,” 1999, preprint available at http://people.csail.mit.edu/rivest/Rivest- commitment.pdf

    work page 1999

  44. [44]

    Precomputing oblivious transfer,

    D. Beaver, “Precomputing oblivious transfer,” in Annual Int. Cryptology Conference. Springer, 1995, pp. 97–109

    work page 1995

  45. [45]

    A two-party protocol with trusted initializer for computing the inner prod- uct,

    R. Dowsley, J. Van De Graaf, D. Marques, and A. C. Nascimento, “A two-party protocol with trusted initializer for computing the inner prod- uct,” in International Workshop on Information Security Applications . Springer, 2010, pp. 337–350

    work page 2010

  46. [46]

    Universally composable and statistically secure verifiable secret sharing scheme based on pre-distributed data,

    R. Dowsley, J. M ¨uller-Quade, A. Otsuka, G. Hanaoka, H. Imai, and A. C. A. Nascimento, “Universally composable and statistically secure verifiable secret sharing scheme based on pre-distributed data,” IEICE Transactions, vol. 94-A, no. 2, pp. 725–734, 2011

    work page 2011

  47. [47]

    On the power of correlated randomness in secure com- putation,

    Y . Ishai, E. Kushilevitz, S. Meldgaard, C. Orlandi, and A. Paskin- Cherniavsky, “On the power of correlated randomness in secure com- putation,” in Theory of Cryptography . Springer, 2013, pp. 600–620

    work page 2013

  48. [48]

    Information-theoretically se- cure oblivious polynomial evaluation in the commodity-based model,

    R. Tonicelli, A. C. A. Nascimento, R. Dowsley, J. M ¨uller-Quade, H. Imai, G. Hanaoka, and A. Otsuka, “Information-theoretically se- cure oblivious polynomial evaluation in the commodity-based model,” Int. Journal of Information Security , vol. 14, no. 1, pp. 73–84, 2015

    work page 2015

  49. [49]

    Unconditionally secure, universally composable privacy preserving linear algebra,

    B. David, R. Dowsley, J. van de Graaf, D. Marques, A. C. A. Nascimento, and A. C. B. Pinto, “Unconditionally secure, universally composable privacy preserving linear algebra,” IEEE Transactions on Information Forensics and Security , vol. 11, no. 1, pp. 59–73, 2016

    work page 2016

  50. [50]

    Efficient multiparty protocols using circuit randomization,

    D. Beaver, “Efficient multiparty protocols using circuit randomization,” in Annual Int. Cryptology Conf. Springer, 1991, pp. 420–432

    work page 1991

  51. [51]

    Cryptography based on correlated data: Foundations and practice,

    R. Dowsley, “Cryptography based on correlated data: Foundations and practice,” Ph.D. dissertation, Karlsruhe Institute of Technology, Germany, 2016

    work page 2016

  52. [52]

    Secure computation with fixed-point num- bers,

    O. Catrina and A. Saxena, “Secure computation with fixed-point num- bers,” in Int. Conf. on Financial Cryptography and Data Security . Springer, 2010, pp. 35–50

    work page 2010

  53. [53]

    Selective transfer learning for EEG-based drowsiness detection,

    C.-S. Wei, Y .-P. Lin, Y .-T. Wang, T.-P. Jung, N. Bigdely-Shamlo, and C.- T. Lin, “Selective transfer learning for EEG-based drowsiness detection,” in IEEE SMC 2015 , 2015, pp. 3229–3232

    work page 2015

  54. [54]

    Lapse in alertness: coherence of fluctuations in performance and EEG spectrum,

    S. Makeig and M. Inlow, “Lapse in alertness: coherence of fluctuations in performance and EEG spectrum,” Electroencephalography and Clinical Neurophysiology, vol. 86, no. 1, pp. 23–35, 1993

    work page 1993