Aviv A. Rosenberg

Currently a machine learning engineer and data scientist at Sibylla, a fintech startup, where I lead the design and implementation of our machine-learning-based algorithmic trading platform.

Prior to this role, I completed my PhD in the Technion Computer Science Faculty, advised by Prof. Alex Bronstein. I conducted multidisciplinary research focusing on machine learning and statistical tools, with applications in medicine and biology. Some research highlights include: Non-linear and scalable extension for vector quantile regression, a powerful tool allowing general statistical inference by modelling conditional quantiles; Distributional analysis and hypothesis testing for codon-specific protein backbone angles; computationally identifying protein structures defying the “one sequence, one structure” principle; Deep-learning based ECG analysis system focused on clinical applicability and statistical performance. During that time I was also head TA of the CS faculty’s Deep Learning course (2019-2022). I also hold an MSc in Biomedical Engineering and a BSc in Electrical Engineering, both from the Technion.

In addition to my research pursuits, I have a strong technical background in software engineering. Before starting my PhD I worked in multiple technical roles. I have extensive experience writing robust, testable, well-designed code, leading software teams and architecting large-scale systems.

news

Oct 27, 2023	New paper in PNAS: Our new paper in the domain of computational structural biology was recently published in the Proceedings of the National Academy of Sciences. We performed distributional analysis of protein backbone angles using a novel a cross-peptide-bond Ramachandran plot that captures the conformational preferences of the amino acid pairs, and demonstrate that it conveys biologically meaningful information, not apparent in the traditional Ramachandran plot.
Jul 1, 2023	New workshop papers in ICML: Two of our new papers have been accepted to the ICML 2023 workshop Frontiers4LCD. In the first paper we propose a novel approach for estimating high-dimensional conditional quantile functions on manifolds. In the second paper we propose a novel continuous formulation of vector quantile regression that allows for accurate, scalable, differentiable, and invertible estimation of nonlinear conditional vector quantile functions.
May 22, 2023	PhD defense completed: I have successfully defended by PhD thesis, as part of my doctoral degree requirements. I would like to sincerely thank the committee members, Prof. Ron Kimmel (Technion), Prof. Joel L. Sussman (Weizmann) and Prof. Mickey Scheinowitz (TAU) for the interesting and engaging discussion.
Feb 1, 2023	New paper in ICLR: Our paper, “Fast Nonlinear Vector Quantile Regression” has been published in the International Conference for Learning Representations (ICLR 2023). In this work, we extend Vector Quantile Regression to support non-linear specification, while ensuring monotonicity and scaling to millions of samples.
Dec 20, 2022	New paper in Scientific Reports: Our recent work, has been published. Building on our previous work, we apply machine learning approaches to demonstrate that protein structures carry information about their genetic coding.
Jun 6, 2022	Featured: Our recent work about the association between protein structure and synonymous genetic coding was featured on the Technion website! See english and hebrew.
May 20, 2022	New paper in Nature Communications: Our paper, “Codon-specific Ramachandran plots show amino acid backbone conformation depends on identity of the translated codon” has been published in Nature Communications. In this work we applied powerful statistical methods to uncover novel associations between synonymous genetic coding and protein structure.
Dec 8, 2021	Scholarship: I have been awarded with the Gutwirth Excellence Scholarship based on my PhD research. This scholarship is intended for PhD students at Technion, based on academic excellence and research achievements, where priority is given to students whose academic achievements indicate a high potential for further career as independent researchers.
Jul 23, 2021	Featured: Our paper about clinically-relevant deep-learning-based ECG analysis was featured on the Technion website! See english and hebrew.
Jun 14, 2021	New paper in PNAS: Our recent work paper was published in PNAS. We tackle the issues of real-world clinical applicability of deep-learning based ECG analysis.
Aug 24, 2020	Scholarship: I have been awarded the Technion Machine Learning and Intelligent Systems (MLIS) Scholarship (funded by the Council for Higher Education) for outstanding research in multi-disciplinary Data Science.
Jul 19, 2020	New paper: Our new paper was published in Nature Scientific Reports.

selected publications

2023

PNAS

An amino domino model described by a cross peptide bond Ramachandran plot defines amino acid pairs as local structural units

Aviv A. Rosenberg, Nitsan YehiShalom, Ailie Marx, and Alex M Bronstein

Proceedings of the National Academy of Sciences, 2023

Abs HTML PDF

Protein structure, both at the global and local level, dictates function. Proteins fold from chains of amino acids, forming secondary structures, α-helices and β-strands, that, at least for globular proteins, subsequently fold into a three-dimensional structure. Here, we show that a Ramachandran-type plot focusing on the two dihedral angles separated by the peptide bond, and entirely contained within an amino acid pair, defines a local structural unit. We further demonstrate the usefulness of this cross-peptide-bond Ramachandran plot by showing that it captures β-turn conformations in coil regions, that traditional Ramachandran plot outliers fall into occupied regions of our plot, and that thermophilic proteins prefer specific amino acid pair conformations. Further, we demonstrate experimentally that the effect of a point mutation on backbone conformation and protein stability depends on the amino acid pair context, i.e., the identity of the adjacent amino acid, in a manner predictable by our method.
ICLR

Fast Nonlinear Vector Quantile Regression

Aviv A. Rosenberg, Sanketh Vedula, Yaniv Romano, and Alex M. Bronstein

In The Eleventh International Conference on Learning Representations , 2023

Abs HTML PDF

Quantile regression (QR) is a powerful tool for estimating one or more conditional quantiles of a target variable Y given explanatory features X. A limitation of QR is that it is only defined for scalar target variables, due to the formulation of its objective function, and since the notion of quantiles has no standard definition for multivariate distributions. Recently, vector quantile regression (VQR) was proposed as an extension of QR for vector-valued target variables, thanks to a meaningful generalization of the notion of quantiles to multivariate distributions via optimal transport. Despite its elegance, VQR is arguably not applicable in practice due to several limitations: (i) it assumes a linear model for the quantiles of the target Y given the features X; (ii) its exact formulation is intractable even for modestly-sized problems in terms of target dimensions, number of regressed quantile levels, or number of features, and its relaxed dual formulation may violate the monotonicity of the estimated quantiles; (iii) no fast or scalable solvers for VQR currently exist. In this work we fully address these limitations, namely: (i) We extend VQR to the non-linear case, showing substantial improvement over linear VQR; (ii) We propose vector monotone rearrangement, a method which ensures the quantile functions estimated by VQR are monotone functions; (iii) We provide fast, GPU-accelerated solvers for linear and nonlinear VQR which maintain a fixed memory footprint, and demonstrate that they scale to millions of samples and thousands of quantile levels; (iv) We release an optimized python package of our solvers as to widespread the use of VQR in real-world applications.

2022

NatComm

Codon-specific Ramachandran plots show amino acid backbone conformation depends on identity of the translated codon

Aviv A. Rosenberg, Ailie Marx, and Alex M. Bronstein

Nature Communications, May 2022

Abs HTML PDF

Synonymous codons translate into chemically identical amino acids. Once considered inconsequential to the formation of the protein product, there is evidence to suggest that codon usage affects co-translational protein folding and the final structure of the expressed protein. Here we develop a method for computing and comparing codon-specific Ramachandran plots and demonstrate that the backbone dihedral angle distributions of some synonymous codons are distinguishable with statistical significance for some secondary structures. This shows that there exists a dependence between codon identity and backbone torsion of the translated amino acid. Although these findings cannot pinpoint the causal direction of this dependence, we discuss the vast biological implications should coding be shown to directly shape protein conformation and demonstrate the usefulness of this method as a tool for probing associations between codon usage and protein structure. Finally, we urge for the inclusion of exact genetic information into structural databases.

2021

PNAS

Meeting the unmet needs of clinicians from AI systems showcased for cardiology with deep-learning–based ECG analysis

Yonatan Elul, Aviv A. Rosenberg, Assaf Schuster, Alex M. Bronstein, and Yael Yaniv

Proceedings of the National Academy of Sciences, May 2021

Abs HTML PDF

The use of artificial intelligence (AI) in medicine, particularly deep learning, has gained considerable attention recently. Although some works boast superior capabilities compared to clinicians, actual deployments of AI systems in the clinic are scarce. We describe four important gaps on the machine-learning side responsible for this discrepancy by first formulating them in a way that is actionable by AI researchers and then systematically addressing these needs. Aiming beyond the search for better model architectures or improved accuracy, we focus directly on the challenges of clinical usefulness as stated by medical professionals in the literature. Our results show that deep-learning systems can be robust, trustworthy, explainable, and transparent while retaining the superior level of performance these algorithms are known for.Despite their great promise, artificial intelligence (AI) systems have yet to become ubiquitous in the daily practice of medicine largely due to several crucial unmet needs of healthcare practitioners. These include lack of explanations in clinically meaningful terms, handling the presence of unknown medical conditions, and transparency regarding the system’s limitations, both in terms of statistical performance as well as recognizing situations for which the system’s predictions are irrelevant. We articulate these unmet clinical needs as machine-learning (ML) problems and systematically address them with cutting-edge ML techniques. We focus on electrocardiogram (ECG) analysis as an example domain in which AI has great potential and tackle two challenging tasks: the detection of a heterogeneous mix of known and unknown arrhythmias from ECG and the identification of underlying cardio-pathology from segments annotated as normal sinus rhythm recorded in patients with an intermittent arrhythmia. We validate our methods by simulating a screening for arrhythmias in a large-scale population while adhering to statistical significance requirements. Specifically, our system 1) visualizes the relative importance of each part of an ECG segment for the final model decision; 2) upholds specified statistical constraints on its out-of-sample performance and provides uncertainty estimation for its predictions; 3) handles inputs containing unknown rhythm types; and 4) handles data from unseen patients while also flagging cases in which the model’s outputs are not usable for a specific patient. This work represents a significant step toward overcoming the limitations currently impeding the integration of AI into clinical practice in cardiology and medicine in general.All study data are included in the article and/or SI Appendix.