Arrow Research search

Author name cluster

Lior Rokach

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

14 papers
2 author rows

Possible papers

14

ECAI Conference 2025 Conference Paper

DiFair-LLM: Evaluating Fairness Disparities in LLMs Toward Demographic Groups

  • Nurit Cohen-Inger
  • Roei Zaady
  • Adir Solomon
  • Lior Rokach
  • Bracha Shapira

Large Language Models (LLMs) are increasingly integrated into real-world applications, making equitable treatment of all demographic groups a critical concern. Existing fairness evaluations often rely on binary, template-based tests, which overlook subtle disparities in open-ended responses. We present DiFair-LLM, a model-agnostic framework for detecting and quantifying fairness disparities - any unequal treatment that benefits or disadvantages a demographic group. DiFair-LLM uses open-ended, group-specific and neutral prompts, measures semantic distances between groups’ responses, applies non-parametric statistical tests, and ranks groups by deviation from a neutral baseline. Evaluations across eight state-of-the-art LLMs and multiple demographic attributes reveal minimal disparities for gender but significant differences for age, especially older adults, and ethnicity, with the largest gaps affecting certain non-Caucasian groups. By mapping nuanced patterns of differential treatment rather than flagging only overt bias, DiFair-LLM offers a practical, reproducible approach for auditing fairness and guiding more inclusive LLM deployments.

Details

JBHI Journal 2025 Journal Article

Temporal Integrative Machine Learning for Early Detection of Diabetic Retinopathy Using Fundus Imaging and Electronic Health Records

  • Shvat Messica
  • Seffi Cohen
  • Aviel Hadad
  • Michal Gordon
  • Or Katz
  • Dan Presil
  • Noa Dagan
  • Erez Tsumi

Diabetic Retinopathy (DR), a prevalent diabetes complication leading to blindness, often goes undetected until late stages due to patients seeking help only when symptoms manifest and limited experts' availability. To address these challenges, we present a novel temporal integrative machine learning system that harnesses both fundus images and electronic health records (EHR) for early and enhanced DR detection. Our system uniquely processes EHR data by focusing on temporal trends and long-term patient histories, creating thousands of temporal features that capture their evolving dynamics over time and deliver unparalleled model finesse. This dual-model system includes a temporal tabular model that relies solely on historical medical records and a deep learning multi-modal model that combines these records with fundus images. The models were trained and tested using real clinical data from 5, 000 patients at Soroka Hospital in Israel, comprising 25, 000 retinal images collected over 8 years and electronic health records spanning up to 20 years. Given the primarily unlabeled nature of the data, the training phase employed a pseudo-labeling technique. The models were evaluated and verified by a retina specialist, surpassing existing models with AUROC scores of 0. 881 for the temporal-trend EHR model and 0. 988 for the multi-modal imaging + EHR model. The integration of historical temporal medical data with imaging offers a more dynamic and comprehensive machine-learning system, enhancing DR detection and offering new insights into associated risk factors. This system not only aids physicians in obtaining a holistic view of a patient's health over time but also facilitates fast identification of individuals at high risk for DR.

Details DOI

ECAI Conference 2024 Conference Paper

FairUS - UpSampling Optimized Method for Boosting Fairness

  • Nurit Cohen-Inger
  • Guy Rozenblatt
  • Seffi Cohen
  • Lior Rokach
  • Bracha Shapira

The increasing application of machine learning (ML) in critical areas such as healthcare and finance highlights the importance of fairness in ML models, challenged by biases in training data that can lead to discrimination. We introduce ‘FairUS’, a novel pre-processing method for reducing bias in ML models utilizing the Conditional Generative Adversarial Network (CTGAN) to synthesize upsampled data. Unlike traditional approaches that focus solely on balancing subgroup sample sizes, FairUS strategically optimizes the quantity of synthesized data. This optimization aims to achieve an ideal balance between enhancing fairness and maintaining the overall performance of the model. Extensive evaluations of our method over several canonical datasets show that the proposed method enhances fairness by 2. 7 times more than the related work and 4 times more than the baseline without mitigation, while preserving the performance of the ML model. Moreover, less than a third of the amount of synthetic data was needed on average. Uniquely, the proposed method enables decision-makers to choose the working point between improved fairness and model’s performance according to their preferences.

Details

AAAI Conference 2024 Conference Paper

TTTS: Tree Test Time Simulation for Enhancing Decision Tree Robustness against Adversarial Examples

  • Seffi Cohen
  • Ofir Arbili
  • Yisroel Mirsky
  • Lior Rokach

Decision trees are widely used for addressing learning tasks involving tabular data. Yet, they are susceptible to adversarial attacks. In this paper, we present Tree Test Time Simulation (TTTS), a novel inference-time methodology that incorporates Monte Carlo simulations into decision trees to enhance their robustness. TTTS introduces a probabilistic modification to the decision path, without altering the underlying tree structure. Our comprehensive empirical analysis of 50 datasets yields promising results. Without the presence of any attacks, TTTS has successfully improved model performance from an AUC of 0.714 to 0.773. Under the challenging conditions of white-box attacks, TTTS demonstrated its robustness by boosting performance from an AUC of 0.337 to 0.680. Even when subjected to black-box attacks, TTTS maintains high accuracy and enhances the model's performance from an AUC of 0.628 to 0.719. Compared to defenses such as Feature Squeezing, TTTS proves to be much more effective. We also found that TTTS exhibits similar robustness in decision forest settings across different attacks.

PDF Details DOI

AIIM Journal 2023 Journal Article

Vulnerability of pangolin SARS-CoV-2 lineage assignment to adversarial attack

  • Amiel Meiseles
  • Yair Motro
  • Lior Rokach
  • Jacob Moran-Gilad

Pangolin is the most popular tool for SARS-CoV-2 lineage assignment. During COVID-19, healthcare professionals and policymakers required accurate and timely lineage assignment of SARS-CoV-2 genomes for pandemic response. Therefore, tools such as Pangolin use a machine learning model, pangoLEARN, for fast and accurate lineage assignment. Unfortunately, machine learning models are susceptible to adversarial attacks, in which minute changes to the inputs cause substantial changes in the model prediction. We present an attack that uses the pangoLEARN architecture to find perturbations that change the lineage assignment, often with only 2–3 base pair changes. The attacks we carried out show that pangolin is vulnerable to adversarial attack, with success rates between 0. 98 and 1 for sequences from non-VoC lineages when pangoLEARN is used for lineage assignment. The attacks we carried out are almost never successful against VoC lineages because pangolin uses Usher and Scorpio – the non-machine-learning alternative methods for VoC lineage assignment. A malicious agent could use the proposed attack to fake or mask outbreaks or circulating lineages. Developers of software in the field of microbial genomics should be aware of the vulnerabilities of machine learning based models and mitigate such risks.

Details DOI

JMLR Journal 2022 Journal Article

Deepchecks: A Library for Testing and Validating Machine Learning Models and Data

  • Shir Chorev
  • Philip Tannor
  • Dan Ben Israel
  • Noam Bressler
  • Itay Gabbay
  • Nir Hutnik
  • Jonatan Liberman
  • Matan Perlmutter

This paper presents Deepchecks, a Python library for comprehensively validating machine learning models and data. Our goal is to provide an easy-to-use library comprising many checks related to various issues, such as model predictive performance, data integrity, data distribution mismatches, and more. The package is distributed under the GNU Affero General Public License and relies on core libraries from the scientific Python ecosystem: scikit-learn, PyTorch, NumPy, pandas, and SciPy. [abs] [ pdf ][ bib ] [ code ] &copy JMLR 2022. ( edit, beta )

PDF Details

AAAI Conference 2022 Conference Paper

Q-Ball: Modeling Basketball Games Using Deep Reinforcement Learning

  • Chen Yanai
  • Adir Solomon
  • Gilad Katz
  • Bracha Shapira
  • Lior Rokach

Basketball is one of the most popular types of sports in the world. Recent technological developments have made it possible to collect large amounts of data on the game, analyze it, and discover new insights. We propose a novel approach for modeling basketball games using deep reinforcement learning. By analyzing multiple aspects of both the players and the game, we are able to model the latent connections among players’ movements, actions, and performance, into a single measure – the Q-Ball. Using Q-Ball, we are able to assign scores to the performance of both players and whole teams. Our approach has multiple practical applications, including evaluating and improving players’ game decisions and producing tactical recommendations. We train and evaluate our approach on a large dataset of National Basketball Association games, and show that the Q-Ball is capable of accurately assessing the performance of players and teams. Furthermore, we show that Q-Ball is highly effective in recommending alternatives to players’ actions.

PDF Details

IJCAI Conference 2019 Conference Paper

AugBoost: Gradient Boosting Enhanced with Step-Wise Feature Augmentation

  • Philip Tannor
  • Lior Rokach

Gradient Boosted Decision Trees (GBDT) is a widely used machine learning algorithm, which obtains state-of-the-art results on many machine learning tasks. In this paper we introduce a method for obtaining better results, by augmenting the features in the dataset between the iterations of GBDT. We explore a number of augmentation methods: training an Artificial Neural Network (ANN) and extracting features from it's last hidden layer (supervised), and rotating the feature-space using unsupervised methods such as PCA or Random Projection (RP). These variations on GBDT were tested on 20 classification tasks, on which all of them outperformed GBDT and previous related work.

PDF Details

EAAI Journal 2019 Journal Article

Choosing the right word: Using bidirectional LSTM tagger for writing support systems

  • Victor Makarenkov
  • Lior Rokach
  • Bracha Shapira

Scientific writing is difficult. It is even harder for those for whom English is a second language (ESL learners). Scholars around the world spend a significant amount of time and resources proofreading their work before submitting it for review or publication. In this paper we present a novel machine learning based application for proper word choice task. Proper word choice is a generalization the lexical substitution (LS) and grammatical error correction (GEC) tasks. We demonstrate and evaluate the usefulness of applying bidirectional Long Short Term Memory (LSTM) tagger, for this task. While state-of-the-art grammatical error correction uses error-specific classifiers and machine translation methods, we demonstrate an unsupervised method that is based solely on a high quality text corpus and does not require manually annotated data. We use a bidirectional Recurrent Neural Network (RNN) with LSTM for learning the proper word choice based on a word’s sentential context. We demonstrate and evaluate our application in various settings, including both a domain-specific (scientific), writing task and a general-purpose writing task. We perform both strict machine and human evaluation. We show that our domain-specific and general-purpose models outperform state-of-the-art general context learning. As an additional contribution of this research, we also share our code, pre-trained models, and a new ESL learner test set with the research community.

Details DOI

TIST Journal 2016 Journal Article

Anytime Algorithms for Recommendation Service Providers

  • David Ben-Shimon
  • Lior Rokach
  • Guy Shani
  • Bracha Shapira

Recommender systems (RS) can now be found in many commercial Web sites, often presenting customers with a short list of additional products that they might purchase. Many commercial sites do not typically have the ability and resources to develop their own system and may outsource the RS to a third party. This had led to the growth of a recommendation as a service industry, where companies, referred to as RS providers, provide recommendation services. These companies must carefully balance the cost of building recommendation models and the payment received from the e-business, as these payments are expected to be low. In such a setting, restricting the computational time required for model building is critical for the RS provider to be profitable. In this article, we propose anytime algorithms as an attractive method for balancing computational time and the recommendation model performance, thus tackling the RS provider problem. In an anytime setting, an algorithm can be stopped after any amount of computational time, always ensuring that a valid, although suboptimal, solution will be returned. Given sufficient time, however, the algorithm should converge to an optimal solution. In this setting, it is important to evaluate the quality of the returned solution over time, monitoring quality improvement. This is significantly different from traditional evaluation methods, which mostly estimate the performance of the algorithm only after its convergence is given sufficient time. We show that the popular item-item top-N recommendation approach can be brought into the anytime framework by smartly considering the order by which item pairs are being evaluated. We experimentally show that the time-accuracy trade-off can be significantly improved for this specific problem.

Details DOI

TIST Journal 2013 Journal Article

Computationally efficient link prediction in a variety of social networks

  • Michael Fire
  • Lena Tenenboim-Chekina
  • Rami Puzis
  • Ofrit Lesser
  • Lior Rokach
  • Yuval Elovici

Online social networking sites have become increasingly popular over the last few years. As a result, new interdisciplinary research directions have emerged in which social network analysis methods are applied to networks containing hundreds of millions of users. Unfortunately, links between individuals may be missing either due to an imperfect acquirement process or because they are not yet reflected in the online network (i.e., friends in the real world did not form a virtual connection). The primary bottleneck in link prediction techniques is extracting the structural features required for classifying links. In this article, we propose a set of simple, easy-to-compute structural features that can be analyzed to identify missing links. We show that by using simple structural features, a machine learning classifier can successfully identify missing links, even when applied to a predicament of classifying links between individuals with at least one common friend. We also present a method for calculating the amount of data needed in order to build more accurate classifiers. The new Friends measure and Same community features we developed are shown to be good predictors for missing links. An evaluation experiment was performed on ten large social networks datasets: Academia.edu, DBLP, Facebook, Flickr, Flixster, Google+, Gowalla, TheMarker, Twitter, and YouTube. Our methods can provide social network site operators with the capability of helping users to find known, offline contacts and to discover new friends online. They may also be used for exposing hidden links in online social networks.

Details DOI

IJCAI Conference 2013 Conference Paper

Machine-Learning-Based Circuit Synthesis

  • Lior Rokach
  • Meir Kalech
  • Gregory Provan
  • Alexander Feldman

Multi-level logic synthesis is a problem of immense practical significance, and is a key to developing circuits that optimize a number of parameters, such as depth, energy dissipation, reliability, etc. The problem can be defined as the task of taking a collection of components from which one wants to synthesize a circuit that optimizes a particular objective function. This problem is computationally hard, and there are very few automated approaches for its solution. To solve this problem we propose an algorithm, called Circuit-Decomposition Engine (CDE), that is based on learning decision trees, and uses a greedy approach for function learning. We empirically demonstrate that CDE, when given a library of different component types, can learn the function of Disjunctive Normal Form (DNF) Boolean representations and synthesize circuit structure using the input library. We compare the structure of the synthesized circuits with that of well-known circuits using a range of circuit similarity metrics.

PDF Details DOI

JMLR Journal 2012 Journal Article

Mal-ID: Automatic Malware Detection Using Common Segment Analysis and Meta-Features

  • Gil Tahan
  • Lior Rokach
  • Yuval Shahar

This paper proposes several novel methods, based on machine learning, to detect malware in executable files without any need for preprocessing, such as unpacking or disassembling. The basic method (Mal-ID) is a new static (form-based) analysis methodology that uses common segment analysis in order to detect malware files. By using common segment analysis, Mal-ID is able to discard malware parts that originate from benign code. In addition, Mal-ID uses a new kind of feature, termed meta-feature, to better capture the properties of the analyzed segments. Rather than using the entire file, as is usually the case with machine learning based techniques, the new approach detects malware on the segment level. This study also introduces two Mal-ID extensions that improve the Mal-ID basic method in various aspects. We rigorously evaluated Mal-ID and its two extensions with more than ten performance measures, and compared them to the highly rated boosted decision tree method under identical settings. The evaluation demonstrated that Mal-ID and the two Mal-ID extensions outperformed the boosted decision tree method in almost all respects. In addition, the results indicated that by extracting meaningful features, it is sufficient to employ one simple detection rule for classifying executable files. [abs] [ pdf ][ bib ] &copy JMLR 2012. ( edit, beta )

PDF Details
v2026.09.13