Search

Your search keyword '"Rodríguez-Henriquez P"' showing total 6 results
Start Over Author "Rodríguez-Henriquez P" Search Limiters Full Text
6 results on '"Rodríguez-Henriquez P"'

1. Polynomial Time Cryptanalytic Extraction of Deep Neural Networks in the Hard-Label Setting

Author

Carlini, Nicholas, Chávez-Saab, Jorge, Hambitzer, Anna, Rodríguez-Henríquez, Francisco, and Shamir, Adi

Subjects
Computer Science - Cryptography and Security, Computer Science - Artificial Intelligence
Abstract

Deep neural networks (DNNs) are valuable assets, yet their public accessibility raises security concerns about parameter extraction by malicious actors. Recent work by Carlini et al. (crypto'20) and Canales-Mart\'inez et al. (eurocrypt'24) has drawn parallels between this issue and block cipher key extraction via chosen plaintext attacks. Leveraging differential cryptanalysis, they demonstrated that all the weights and biases of black-box ReLU-based DNNs could be inferred using a polynomial number of queries and computational time. However, their attacks relied on the availability of the exact numeric value of output logits, which allowed the calculation of their derivatives. To overcome this limitation, Chen et al. (asiacrypt'24) tackled the more realistic hard-label scenario, where only the final classification label (e.g., "dog" or "car") is accessible to the attacker. They proposed an extraction method requiring a polynomial number of queries but an exponential execution time. In addition, their approach was applicable only to a restricted set of architectures, could deal only with binary classifiers, and was demonstrated only on tiny neural networks with up to four neurons split among up to two hidden layers. This paper introduces new techniques that, for the first time, achieve cryptanalytic extraction of DNN parameters in the most challenging hard-label setting, using both a polynomial number of queries and polynomial time. We validate our approach by extracting nearly one million parameters from a DNN trained on the CIFAR-10 dataset, comprising 832 neurons in four hidden layers. Our results reveal the surprising fact that all the weights of a ReLU-based DNN can be efficiently determined by analyzing only the geometric shape of its decision boundaries.

Published
2024

2. Ubiquitous broad-line emission and the relation between ionized gas outflows and Lyman continuum escape in Green Pea galaxies

Author

Amorín, R. O., Rodríguez-Henríquez, M., Fernández, V., Vílchez, J. M., Marques-Chaves, R., Schaerer, D., Izotov, Y. I., Firpo, V., Guseva, N., Jaskot, A. E., Komarova, L., Muñoz-Vergara, D., Oey, M. S., Bait, O., Carr, C., Chisholm, J., Ferguson, H., Flury, S. R., Giavalisco, M., Hayes, M. J., Henry, A., Ji, Z., King, W., Leclercq, F., Östlin, G., Pentericci, L., Saldana-Lopez, A., Thuan, T. X., Trebitsch, M., Wang, B., Worseck, G., and Xu, X.

Subjects
Astrophysics - Astrophysics of Galaxies, Astrophysics - Cosmology and Nongalactic Astrophysics
Abstract

We report observational evidence of highly turbulent ionized gas kinematics in a sample of 20 Lyman continuum (LyC) emitters (LCEs) at low redshift ($z\sim 0.3$). Detailed Gaussian modeling of optical emission line profiles in high-dispersion spectra consistently shows that both bright recombination and collisionally excited lines can be fitted as one or two narrow components with intrinsic velocity dispersion of $\sigma$ $\sim$ 40-100 km s$^{-1}$, in addition to a broader component with $\sigma \sim$ 100-300 km s$^{-1}$, which contributes up to $\sim$40% of the total flux and is preferentially blueshifted from the systemic velocity. We interpret the narrow emission as highly ionized gas close to the young massive star clusters and the broader emission as a signpost of unresolved ionized outflows, resulting from massive stars and supernova feedback. We find a significant correlation between the width of the broad emission and the LyC escape fraction, with strong LCEs exhibiting more complex and broader line profiles than galaxies with weaker or undetected LyC emission. We provide new observational evidence supporting predictions from models and simulations; our findings suggest that gas turbulence and outflows resulting from strong radiative and mechanical feedback play a key role in clearing channels through which LyC photons escape from galaxies. We propose that the detection of blueshifted broad emission in the nebular lines of compact extreme emission-line galaxies can provide a new indirect diagnostic of Lyman photon escape, which could be useful to identify potential LyC leakers in the epoch of reionization with the JWST., Comment: 11 Pages, 7 Figures. A&A in Press

Published
2024
Full Text
View/download PDF

3. Faulty isogenies: a new kind of leakage

Author

Adj, Gora, Chi-Domínguez, Jesús-Javier, Mateu, Víctor, and Rodríguez-Henríquez, Francisco

Subjects
Mathematics - Number Theory, Computer Science - Cryptography and Security
Abstract

In SIDH and SIKE protocols, public keys are defined over quadratic extensions of prime fields. We present in this work a projective invariant property characterizing affine Montgomery curves defined over prime fields. We then force a secret 3-isogeny chain to repeatedly pass through a curve defined over a prime field in order to exploit the new property and inject zeros in the A-coefficient of an intermediate curve to successfully recover the isogeny chain one step at a time. Our results introduce a new kind of fault attacks applicable to SIDH and SIKE., Comment: Update: extended countermeasures

Published
2022

4. Extending the GLS endomorphism to speed up GHS Weil descent using Magma

Author

Chi-Domínguez, Jesús-Javier, Rodríguez-Henríquez, Francisco, and Smith, Benjamin

Subjects
Computer Science - Cryptography and Security
Abstract

Let $q = 2^n$, and let $E / \mathbb{F}_{q^{\ell}}$ be a generalized Galbraith--Lin--Scott (GLS) binary curve, with $\ell \ge 2$ and $(\ell, n) = 1$.We show that the GLS endomorphism on $E / \mathbb{F}_{q^{\ell}}$ induces an efficient endomorphism on the Jacobian $J_H(\mathbb{F}_q)$ of the genus-$g$ hyperelliptic curve $H$ corresponding to the image of the GHS Weil-descent attack applied to $E/\mathbb{F}_{q^\ell}$, and that this endomorphism yields a factor-$n$ speedup when using standard index-calculus procedures for solving the Discrete Logarithm Problem (DLP) on $J_H(\mathbb{F}_q)$. Our analysis is backed up by the explicit computation of a discrete logarithm defined on a prime-order subgroup of a GLS elliptic curve over the field $\mathbb{F}_{2^{5\cdot 31}}$. A Magma implementation of our algorithm finds the aforementioned discrete logarithm in about $1,035$ CPU-days., Comment: Finite Fields and Their Applications, Elsevier, In press, 75

Published
2021
Full Text
View/download PDF

5. Stronger and Faster Side-Channel Protections for CSIDH

Author

Cervantes-Vázquez, Daniel, Chenu, Mathilde, Chi-Domínguez, Jesús-Javier, De Feo, Luca, Rodríguez-Henríquez, Francisco, and Smith, Benjamin

Subjects
Computer Science - Cryptography and Security
Abstract

CSIDH is a recent quantum-resistant primitive based on the difficulty of finding isogeny paths between supersingular curves. Recently, two constant-time versions of CSIDH have been proposed: first by Meyer, Campos and Reith, and then by Onuki, Aikawa, Yamazaki and Takagi. While both offer protection against timing attacks and simple power consumption analysis, they are vulnerable to more powerful attacks such as fault injections. In this work, we identify and repair two oversights in these algorithms that compromised their constant-time character. By exploiting Edwards arithmetic and optimal addition chains, we produce the fastest constant-time version of CSIDH to date. We then consider the stronger attack scenario of fault injection, which is relevant for the security of CSIDH static keys in embedded hardware. We propose and evaluate a dummy-free CSIDH algorithm. While these CSIDH variants are slower, their performance is still within a small constant factor of less-protected variants. Finally, we discuss derandomized CSIDH algorithms., Comment: This work has been accepted in LATINCRYPT-2019

Published
2019

6. Methotrexate and rheumatoid arthritis associated interstitial lung disease.

Author

Juge PA, Lee JS, Lau J, Kawano-Dourado L, Rojas Serrano J, Sebastiani M, Koduri G, Matteson E, Bonfiglioli K, Sawamura M, Kairalla R, Cavagna L, Bozzalla Cassione E, Manfredi A, Mejia M, Rodríguez-Henriquez P, González-Pérez MI, Falfán-Valencia R, Buendia-Roldán I, Pérez-Rubio G, Ebstein E, Gazal S, Borie R, Ottaviani S, Kannengiesser C, Wallaert B, Uzunhan Y, Nunes H, Valeyre D, Saidenberg-Kermanac'h N, Boissier MC, Wemeau-Stervinou L, Flipo RM, Marchand-Adam S, Richette P, Allanore Y, Dromer C, Truchetet ME, Richez C, Schaeverbeke T, Lioté H, Thabut G, Deane KD, Solomon JJ, Doyle T, Ryu JH, Rosas I, Holers VM, Boileau C, Debray MP, Porcher R, Schwartz DA, Vassallo R, Crestani B, and Dieudé P

Subjects
Case-Control Studies, Humans, Methotrexate adverse effects, Antirheumatic Agents adverse effects, Arthritis, Rheumatoid complications, Arthritis, Rheumatoid drug therapy, Lung Diseases, Interstitial chemically induced, Lung Diseases, Interstitial drug therapy
Abstract

Question Addressed by the Study: Methotrexate (MTX) is a key anchor drug for rheumatoid arthritis (RA) management. Fibrotic interstitial lung disease (ILD) is a common complication of RA. Whether MTX exposure increases the risk of ILD in patients with RA is disputed. We aimed to evaluate the association of prior MTX use with development of RA-ILD., Methods: Through a case-control study design with discovery and international replication samples, we examined the association of MTX exposure with ILD in 410 patients with chronic fibrotic ILD associated with RA (RA-ILD) and 673 patients with RA without ILD. Estimates were pooled over the different samples using meta-analysis techniques., Results: Analysis of the discovery sample revealed an inverse relationship between MTX exposure and RA-ILD (adjusted OR 0.46, 95% CI 0.24-0.90; p=0.022), which was confirmed in the replication samples (pooled adjusted OR 0.39, 95% CI 0.19-0.79; p=0.009). The combined estimate using both the derivation and validation samples revealed an adjusted OR of 0.43 (95% CI 0.26-0.69; p=0.0006). MTX ever-users were less frequent among patients with RA-ILD compared to those without ILD, irrespective of chest high-resolution computed tomography pattern. In patients with RA-ILD, ILD detection was significantly delayed in MTX ever-users compared to never-users (11.4±10.4 years and 4.0±7.4 years, respectively; p<0.001)., Answer to the Question: Our results suggest that MTX use is not associated with an increased risk of RA-ILD in patients with RA, and that ILD was detected later in MTX-treated patients., Competing Interests: Conflict of interest: P-A. Juge has nothing to disclose. Conflict of interest: J.S. Lee reports grants from NIH, personal fees for advisory board work from Genentech and Celgene, outside the submitted work. Conflict of interest: J. Lau has nothing to disclose. Conflict of interest: L. Kawano-Dourado has nothing to disclose. Conflict of interest: J. Rojas-Serrano has nothing to disclose. Conflict of interest: M. Sebastiani has nothing to disclose. Conflict of interest: G. Koduri has nothing to disclose. Conflict of interest: E. Matteson has nothing to disclose. Conflict of interest: K. Bonfiglioli has nothing to disclose. Conflict of interest: M. Sawamura has nothing to disclose. Conflict of interest: R. Kairalla has nothing to disclose. Conflict of interest: L. Cavagna has nothing to disclose. Conflict of interest: E. Bozzalla Cassione has nothing to disclose. Conflict of interest: A. Manfredi has nothing to disclose. Conflict of interest: M. Mejia has nothing to disclose. Conflict of interest: P. Rodríguez-Henriquez has nothing to disclose. Conflict of interest: M.I. González Pérez has nothing to disclose. Conflict of interest: R. Falfán-Valencia has nothing to disclose. Conflict of interest: I. Buendia-Roldán has nothing to disclose. Conflict of interest: G. Pérez-Rubio has nothing to disclose. Conflict of interest: E. Ebstein reports personal fees from Sanofi, outside the submitted work. Conflict of interest: S. Gazal has nothing to disclose. Conflict of interest: R. Borie reports grants and personal fees for lectures from Roche and Boehringer Ingelheim, outside the submitted work. Conflict of interest: S. Ottaviani has nothing to disclose. Conflict of interest: C. Kannengiesser has nothing to disclose. Conflict of interest: B. Wallaert reports grants and personal fees for advisory board work and meeting attendance from Boehringer Ingelheim and Roche, outside the submitted work. Conflict of interest: Y. Uzunhan reports personal fees from Roche and Boehringer Ingelheim, non-financial support from Oxyvie, outside the submitted work. Conflict of interest: H. Nunes has nothing to disclose. Conflict of interest: D. Valeyre reports personal fees for advisory board work from Roche and Boehringer Ingelheim, personal fees for lectures from AstraZeneca, outside the submitted work. Conflict of interest: N. Saidenberg-Kermanac'h has nothing to disclose. Conflict of interest: M-C. Boissier has nothing to disclose. Conflict of interest: L. Wemeau-Stervinou reports personal fees for lectures and travel support from Roche, personal fees for lectures and advisory board work, and travel support from Boehringer-Ingelheim, personal fees for lectures from Janssen-Cilag and Bristol-Myers-Squibb, outside the submitted work. Conflict of interest: R.M. Flipo reports grants and personal fees from Roche Chugai, Abbvie and Pfizer, personal fees from Bristol-Meyers Squibb, outside the submitted work. Conflict of interest: S. Marchand-Adam reports fees for research, lectures, meeting attendance, consultancy and advisory board work from Roche, Boehringer Ingelheim and Novartis, outside the submitted work. Conflict of interest: P. Richette reports personal fees from Ipsen/Menarini, AstraZeneca, Savient and Grünenthal, outside the submitted work. Conflict of interest: Y. Allanore reports personal fees from Actelion, Bayer, Bristol-Myers Squibb, Boehringer and Inventiva, grants from Sanofi and Roche, outside the submitted work. Conflict of interest: C. Dromer has nothing to disclose. Conflict of interest: M-E. Truchetet has nothing to disclose. Conflict of interest: C. Richez has nothing to disclose. Conflict of interest: T. Schaeverbeke has nothing to disclose. Conflict of interest: H. Lioté has nothing to disclose. Conflict of interest: G. Thabut reports personal fees from AstraZeneca, outside the submitted work. Conflict of interest: K.D. Deane has nothing to disclose. Conflict of interest: J. Solomon has nothing to disclose. Conflict of interest: T. Doyle has nothing to disclose. Conflict of interest: J.H. Ryu has nothing to disclose. Conflict of interest: I. Rosas reports personal fees for advisory board work from Genentech, Boehringer and Three Lakes Partners, outside the submitted work. Conflict of interest: V.M. Holers reports grants from NIH/NIAID (U01 Grant), during the conduct of the study. Conflict of interest: C. Boileau has nothing to disclose. Conflict of interest: M-P. Debray reports personal fees and non-financial support for travel to meetings from Boehringer Ingelheim and Roche, outside the submitted work. Conflict of interest: R. Porcher has nothing to disclose. Conflict of interest: D.A. Schwartz reports grants from NIH-NHLBI (P01 HL092870, R01 HL097163, R33 HL120770 and UH2 HL123442) and DOD Focused Program (W81XWH-17-1-0597), during the conduct of the study; personal fees for consultancy and advisory board work from NuMedii, Inc., and is an employee of Eleven P15, Inc., outside the submitted work; and has a patent Compositions and Methods of Treating or Preventing Fibrotic Diseases pending, a patent Biomarkers for the Diagnosis and Treatment of Fibrotic Lung Disease pending, and a patent Methods and Compositions for Risk Prediction, Diagnosis, Prognosis, and Treatment of Pulmonary Disorders issued. Conflict of interest: R. Vassallo reports grants from Pfizer, Bristol-Myers-Squibb and SunPharma, outside the submitted work. Conflict of interest: B. Crestani reports grants from Apellis and MedImmune, grants and personal fees for lectures from Boehringer Ingelheim and Roche, personal fees for lectures from AstraZeneca and Sanofi, outside the submitted work. Conflict of interest: P. Dieudé reports fees for consultancy from Pfizer, Abbvie and MSD, grants and personal fees for consultancy and lectures from Roche, Chugai and BMS, outside the submitted work., (Copyright ©ERS 2021.)

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results