Firuz Kamalov

Firuz Kamalov, Amir F. Atiya, Dina Elreedy

Imbalanced data is a frequently encountered problem in machine learning. Despite a vast amount of literature on sampling techniques for imbalanced data, there is a limited number of studies that address the issue of the optimal sampling ratio. In this paper, we attempt to fill the gap in the literature by conducting a large scale study of the effects of sampling ratio on classification accuracy. We consider 10 popular sampling methods and evaluate their performance over a range of ratios based on 20 datasets. The results of the numerical experiments suggest that the optimal sampling ratio is between 0.7 and 0.8 albeit the exact ratio varies depending on the dataset. Furthermore, we find that while factors such the original imbalance ratio or the number of features do not play a discernible role in determining the optimal ratio, the number of samples in the dataset may have a tangible effect.

Firuz Kamalov, Hana Sulieman, Aswani Kumar Cherukuri

Feature selection is an important and active field of research in machine learning and data science. Our goal in this paper is to propose a collection of synthetic datasets that can be used as a common reference point for feature selection algorithms. Synthetic datasets allow for precise evaluation of selected features and control of the data parameters for comprehensive assessment. The proposed datasets are based on applications from electronics in order to mimic real life scenarios. To illustrate the utility of the proposed data we employ one of the datasets to test several popular feature selection algorithms. The datasets are made publicly available on GitHub and can be used by researchers to evaluate feature selection algorithms.

Firuz Kamalov, Hana Sulieman, Witold Pedrycz

Imbalanced data affects a wide range of machine learning applications, from healthcare to network security. As SMOTE is one of the most popular approaches to addressing this issue, it is imperative to validate it not only empirically but also theoretically. In this paper, we provide a rigorous theoretical analysis of SMOTE's convergence properties. Concretely, we prove that the synthetic random variable Z converges in probability to the underlying random variable X. We further prove a stronger convergence in mean when X is compact. Finally, we show that lower values of the nearest neighbor rank lead to faster convergence offering actionable guidance to practitioners. The theoretical results are supported by numerical experiments using both real-life and synthetic data. Our work provides a foundational understanding that enhances data augmentation techniques beyond imbalanced data scenarios.

Firuz Kamalov, Fadi Thabtah, R. Sivaraj et al.

We introduce path-sampled integrated gradients (PS-IG), a framework that generalizes feature attribution by computing the expected value over baselines sampled along the linear interpolation path. We prove that PS-IG is mathematically equivalent to path-weighted integrated gradients, provided the weighting function matches the cumulative distribution function of the sampling density. This equivalence allows the stochastic expectation to be evaluated via a deterministic Riemann sum, improving the error convergence rate from $O(m^{-1/2})$ to $O(m^{-1})$ for smooth models. Furthermore, we demonstrate analytically that PS-IG functions as a variance-reducing filter against gradient noise - strictly lowering attribution variance by a factor of 1/3 under uniform sampling - while preserving key axiomatic properties such as linearity and implementation invariance.

Abhiruph Chakravarty, Jatin Karthik Tripathy, Sibi Chakkaravarthy S et al.

Quadruped robots are currently used in industrial robotics as mechanical aid to automate several routine tasks. However, presently, the usage of such a robot in a domestic setting is still very much a part of the research. This paper discusses the understanding and virtual simulation of such a robot capable of detecting and understanding human emotions, generating its gait, and responding via sounds and expression on a screen. To this end, we use a combination of reinforcement learning and software engineering concepts to simulate a quadruped robot that can understand emotions, navigate through various terrains and detect sound sources, and respond to emotions using audio-visual feedback. This paper aims to establish the framework of simulating a quadruped robot that is emotionally intelligent and can primarily respond to audio-visual stimuli using motor or audio response. The emotion detection from the speech was not as performant as ERANNs or Zeta Policy learning, still managing an accuracy of 63.5%. The video emotion detection system produced results that are almost at par with the state of the art, with an accuracy of 99.66%. Due to its "on-policy" learning process, the PPO algorithm was extremely rapid to learn, allowing the simulated dog to demonstrate a remarkably seamless gait across the different cadences and variations. This enabled the quadruped robot to respond to generated stimuli, allowing us to conclude that it functions as predicted and satisfies the aim of this work.

Long Phan, Alice Gatti, Ziwen Han et al. · amazon-science, apple-ml

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90\% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai.

Firuz Kamalov, David Santandreu Calonge, Linda Smail et al.

Artificial intelligence (AI) has transformed various aspects of education, with large language models (LLMs) driving advancements in automated tutoring, assessment, and content generation. However, conventional LLMs are constrained by their reliance on static training data, limited adaptability, and lack of reasoning. To address these limitations and foster more sustainable technological practices, AI agents have emerged as a promising new avenue for educational innovation. In this review, we examine agentic workflows in education according to four major paradigms: reflection, planning, tool use, and multi-agent collaboration. We critically analyze the role of AI agents in education through these key design paradigms, exploring their advantages, applications, and challenges. To illustrate the practical potential of agentic systems, we present a proof-of-concept application: a multi-agent framework for automated essay scoring. Preliminary results suggest this agentic approach may offer improved consistency compared to stand-alone LLMs. Our findings highlight the transformative potential of AI agents in educational settings while underscoring the need for further research into their interpretability, trustworthiness, and sustainable impact on pedagogical impact.

Febrian Kurniawan, Gandeva Bayu Satrya, Firuz Kamalov

The enormous demand for seafood products has led to exploitation of marine resources and near-extinction of some species. In particular, overfishing is one the main issues in sustainable marine development. In alignment with the protection of marine resources and sustainable fishing, this study proposes to advance fish classification techniques that support identifying protected fish species using state-of-the-art machine learning. We use a custom modification of the MobileNet model to design a lightweight classifier called M-MobileNet that is capable of running on limited hardware. As part of the study, we compiled a labeled dataset of 37,462 images of fish found in the waters of the Indonesian archipelago. The proposed model is trained on the dataset to classify images of the captured fish into their species and give recommendations on whether they are consumable or not. Our modified MobileNet model uses only 50\% of the top layer parameters with about 42% GTX 860M utility and achieves up to 97% accuracy in fish classification and determining its consumability. Given the limited computing capacity available on many fishing vessels, the proposed model provides a practical solution to on-site fish classification. In addition, synchronized implementation of the proposed model on multiple vessels can supply valuable information about the movement and location of different species of fish.

Firuz Kamalov, Mohmad Al Falasi, Fadi Thabtah

Integrated Gradients (IG) is a widely used attribution method in explainable artificial intelligence (XAI). In this paper, we introduce Path-Weighted Integrated Gradients (PWIG), a generalization of IG that incorporates a customizable weighting function into the attribution integral. This modification allows for targeted emphasis along different segments of the path between a baseline and the input, enabling improved interpretability, noise mitigation, and the detection of path-dependent feature relevance. We establish its theoretical properties and illustrate its utility through experiments on a dementia classification task using the OASIS-1 MRI dataset. Attribution maps generated by PWIG highlight clinically meaningful brain regions associated with various stages of dementia, providing users with sharp and stable explanations. The results suggest that PWIG offers a flexible and theoretically grounded approach for enhancing attribution quality in complex predictive models.

Linda Smail, David Santandreu Calonge, Firuz Kamalov et al.

This research investigates the potential of Artificial Intelligence (AI) models to bridge the knowledge gap in environmental education among university students. By focusing on prominent large language models (LLMs) such as GPT-3.5, GPT-4, GPT-4o, Gemini, Claude Sonnet, and Llama 2, the study assesses their effectiveness in conveying environmental concepts and, consequently, facilitating environmental education. The investigation employs a standardized tool, the Environmental Knowledge Test (EKT-19), supplemented by targeted questions, to evaluate the environmental knowledge of university students in comparison to the responses generated by the AI models. The results of this study suggest that while AI models possess a vast, readily accessible, and valid knowledge base with the potential to empower both students and academic staff, a human discipline specialist in environmental sciences may still be necessary to validate the accuracy of the information provided.

Firuz Kamalov, Ashraf Elnagar

Imbalanced data occurs in a wide range of scenarios. The skewed distribution of the target variable elicits bias in machine learning algorithms. One of the popular methods to combat imbalanced data is to artificially balance the data through resampling. In this paper, we compare the efficacy of a recently proposed kernel density estimation (KDE) sampling technique in the context of artificial neural networks. We benchmark the KDE sampling method against two base sampling techniques and perform comparative experiments using 8 datasets and 3 neural networks architectures. The results show that KDE sampling produces the best performance on 6 out of 8 datasets. However, it must be used with caution on image datasets. We conclude that KDE sampling is capable of significantly improving the performance of neural networks.

Firuz Kamalov, Sherif Moussa, Ziad El Khatib et al.

In this paper, we apply a fusion machine learning method to construct an automatic intrusion detection system. Concretely, we employ the orthogonal variance decomposition technique to identify the relevant features in network traffic data. The selected features are used to build a deep neural network for intrusion detection. The proposed algorithm achieves 100% detection accuracy in identifying DDoS attacks. The test results indicate a great potential of the proposed method.

Firuz Kamalov, Sherif Moussa, Rita Zgheib et al.

In this paper, we analyze existing feature selection methods to identify the key elements of network traffic data that allow intrusion detection. In addition, we propose a new feature selection method that addresses the challenge of considering continuous input features and discrete target values. We show that the proposed method performs well against the benchmark selection methods. We use our findings to develop a highly effective machine learning-based detection systems that achieves 99.9% accuracy in distinguishing between DDoS and benign signals. We believe that our results can be useful to experts who are interested in designing and building automated intrusion detection systems.

Firuz Kamalov, Linda Smail, Ikhlaas Gurrib

In this paper, we compare various approaches to stock price prediction using neural networks. We analyze the performance fully connected, convolutional, and recurrent architectures in predicting the next day value of S&P 500 index based on its previous values. We further expand our analysis by including three different optimization techniques: Stochastic Gradient Descent, Root Mean Square Propagation, and Adaptive Moment Estimation. The numerical experiments reveal that a single layer recurrent neural network with RMSprop optimizer produces optimal results with validation and test Mean Absolute Error of 0.0150 and 0.0148 respectively.

Firuz Kamalov, Linda Smail, Ikhlaas Gurrib

Stock price prediction has been the focus of a large amount of research but an acceptable solution has so far escaped academics. Recent advances in deep learning have motivated researchers to apply neural networks to stock prediction. In this paper, we propose a convolution-based neural network model for predicting the future value of the S&P 500 index. The proposed model is capable of predicting the next-day direction of the index based on the previous values of the index. Experiments show that our model outperforms a number of benchmarks achieving an accuracy rate of over 55%.

Firuz Kamalov, Aswani Cherukuri, Hana Sulieman et al.

The COVID-19 pandemic has galvanized the machine learning community to create new solutions that can help in the fight against the virus. The body of literature related to applications of machine learning and artificial intelligence to COVID-19 is constantly growing. The goal of this article is to present the latest advances in machine learning research applied to COVID-19. We cover four major areas of research: forecasting, medical diagnostics, drug development, and contact tracing. We review and analyze the most successful state of the art studies. In contrast to other existing surveys on the subject, our article presents a high level overview of the current research that is sufficiently detailed to provide an informed insight.

Firuz Kamalov, Dmitry Denisov

Imbalanced class distribution is a common problem in a number of fields including medical diagnostics, fraud detection, and others. It causes bias in classification algorithms leading to poor performance on the minority class data. In this paper, we propose a novel method for balancing the class distribution in data through intelligent resampling of the minority class instances. The proposed method is based on generating new minority instances in the neighborhood of the existing minority points via a gamma distribution. Our method offers a natural and coherent approach to balancing the data. We conduct a comprehensive numerical analysis of the new sampling technique. The experimental results show that the proposed method outperforms the existing state-of-the-art methods for imbalanced data. Concretely, the new sampling technique produces the best results on 12 out of 24 real life as well as synthetic datasets. For comparison, the SMOTE method achieves the top score on only 1 dataset. We conclude that the new technique offers a simple yet effective sampling approach to balance data.

Firuz Kamalov, Ikhlaas Gurrib

Asset value forecasting has always attracted an enormous amount of interest among researchers in quantitative analysis. The advent of modern machine learning models has introduced new tools to tackle this classical problem. In this paper, we apply machine learning algorithms to hitherto unexplored question of forecasting instances of significant fluctuations in currency exchange rates. We perform analysis of nine modern machine learning algorithms using data on four major currency pairs over a 10 year period. A key contribution is the novel use of outlier detection methods for this purpose. Numerical experiments show that outlier detection methods substantially outperform traditional machine learning and finance techniques. In addition, we show that a recently proposed new outlier detection method PKDE produces best overall results. Our findings hold across different currency pairs, significance levels, and time horizons indicating the robustness of the proposed method.

Firuz Kamalov

Stock price prediction is a rich research topic that has attracted interest from various areas of science. The recent success of machine learning in speech and image recognition has prompted researchers to apply these methods to asset price prediction. The majority of literature has been devoted to predicting either the actual asset price or the direction of price movement. In this paper, we study a hitherto little explored question of predicting significant changes in stock price based on previous changes using machine learning algorithms. We are particularly interested in the performance of neural network classifiers in the given context. To this end, we construct and test three neural network models including multi-layer perceptron, convolutional net, and long short term memory net. As benchmark models we use random forest and relative strength index methods. The models are tested using 10-year daily stock price data of four major US public companies. Test results show that predicting significant changes in stock price can be accomplished with a high degree of accuracy. In particular, we obtain substantially better results than similar studies that forecast the direction of price change.

Firuz Kamalov

Existing feature selection methods fail to properly account for interactions between features when evaluating feature subsets. In this paper, we attempt to remedy this issue by using orthogonal variance decomposition to evaluate features. The orthogonality of the decomposition allows us to directly calculate the total contribution of a feature to the output variance. Thus we obtain an efficient algorithm for feature evaluation which takes into account interactions among features. Numerical experiments demonstrate that our method accurately identifies relevant features and improves the accuracy of numerical models.

Firuz Kamalov

Imbalanced response variable distribution is a common occurrence in data science. In fields such as fraud detection, medical diagnostics, system intrusion detection and many others where abnormal behavior is rarely observed the data under study often features disproportionate target class distribution. One common way to combat class imbalance is through resampling the minority class to achieve a more balanced distribution. In this paper, we investigate the performance of the sampling method based on kernel density estimation (KDE). We believe that KDE offers a more natural way of generating new instances of minority class that is less prone to overfitting than other standard sampling techniques. It is based on a well established theory of nonparametric statistical estimation. Numerical experiments show that KDE can outperform other sampling techniques on a range of real life datasets as measured by F1-score and G-mean. The results remain consistent across a number of classification algorithms used in the experiments. Furthermore, the proposed method outperforms the benchmark methods irregardless of the class distribution ratio. We conclude, based on the solid theoretical foundation and strong experimental results, that the proposed method would be a valuable tool in problems involving imbalanced class distribution.

Firuz Kamalov, Ho Hon Leung

High-dimensional data poses unique challenges in outlier detection process. Most of the existing algorithms fail to properly address the issues stemming from a large number of features. In particular, outlier detection algorithms perform poorly on data set of small size with a large number of features. In this paper, we propose a novel outlier detection algorithm based on principal component analysis and kernel density estimation. The proposed method is designed to address the challenges of dealing with high-dimensional data by projecting the original data onto a smaller space and using the innate structure of the data to calculate anomaly scores for each data point. Numerical experiments on synthetic and real-life data show that our method performs well on high-dimensional data. In particular, the proposed method outperforms the benchmark methods as measured by the $F_1$-score. Our method also produces better-than-average execution times compared to the benchmark methods.