arXiv subjects

Sasan Tavakkol

Publications and source records attributed to Sasan Tavakkol.

15 recordsLinked to original sources

arXiv2026

Provably Lossless Acceleration of DNN Mutation Testing via Memoization

Mutation analysis has recently reemerged in the context of deep neural networks (DNNs) as a promising, but notoriously costly, approach for assessing test dataset adequacy. Existing techniques speed up DNN mutation testing through lossy approximations that trade efficiency for mutation score accuracy. This paper introduces Mure, the first provably lossless framework for accelerating DNN mutation testing via memoization. Mure is based on the idea that DNN mutants and the original model share substantial redundant computation, so during mutation testing, it executes only the mutated suffixes of each mutant and reuses the common prefix from the original model, which is computed only once. We give a formal account of memoized mutation testing, and prove that Mure is sound, i.e., it produces results equivalent to exhaustive vanilla mutation testing, and identify basic conditions under which speed-up is guaranteed. We have implemented Mure and evaluated it on 15 DNN models of various architectures, complexities, and sizes ranging from a few thousands to millions of parameters. This provides empirical evidence that Mure reduces the computational cost of mutation testing by 44.54%, on average. We also observed that while state-of-the-art techniques tend to yield higher acceleration (up to 88.97%, on average), they come at the cost of some error in mutation score. We further analyze the effect of mutation generation selection ratio on the effectiveness of Mure and observed predictable reductions in memoization opportunities with increasing the percentage of mutated neurons. We observed that Mure offers more than 20% speed-up even when as high as 5% of the neurons are mutated.

cs.SE↗

arXiv2025

Using Fourier Analysis and Mutant Clustering to Accelerate DNN Mutation Testing

Deep neural network (DNN) mutation analysis is a promising approach to evaluating test set adequacy. Due to the large number of generated mutants that must be tested on large datasets, mutation analysis is costly. In this paper, we present a technique, named DM#, for accelerating DNN mutation testing using Fourier analysis. The key insight is that DNN outputs are real-valued functions suitable for Fourier analysis that can be leveraged to quantify mutant behavior using only a few data points. DM# uses the quantified mutant behavior to cluster the mutants so that the ones with similar behavior fall into the same group. A representative from each group is then selected for testing, and the result of the test, e.g., whether the mutant is killed or survived, is reused for all other mutants represented by the selected mutant, obviating the need for testing other mutants. 14 DNN models of sizes ranging from thousands to millions of parameters, trained on different datasets, are used to evaluate DM# and compare it to several baseline techniques. Our results provide empirical evidence on the effectiveness of DM# in accelerating mutation testing by 28.38%, on average, at the average cost of only 0.72% error in mutation score. Moreover, on average, DM# incurs 11.78, 15.16, and 114.36 times less mutation score error compared to random mutant selection, boundary sample selection, and random sample selection techniques, respectively, while generally offering comparable speed-up.

cs.SE↗

arXiv2025

SYNAPSE-G: Bridging Large Language Models and Graph Learning for Rare Event Classification

Scarcity of labeled data, especially for rare events, hinders training effective machine learning models. This paper proposes SYNAPSE-G (Synthetic Augmentation for Positive Sampling via Expansion on Graphs), a novel pipeline leveraging Large Language Models (LLMs) to generate synthetic training data for rare event classification, addressing the cold-start problem. This synthetic data serve as seeds for semi-supervised label propagation on a similarity graph constructed between the seeds and a large unlabeled dataset. This identifies candidate positive examples, subsequently labeled by an oracle (human or LLM). The expanded dataset then trains/fine-tunes a classifier. We theoretically analyze how the quality (validity and diversity) of the synthetic data impacts the precision and recall of our method. Experiments on the imbalanced SST2 and MHS datasets demonstrate SYNAPSE-G's effectiveness in finding positive labels, outperforming baselines including nearest neighbor search.

cs.LG↗

arXiv2025

Less is More: Adaptive Coverage for Synthetic Training Data

Synthetic training data generation with Large Language Models (LLMs) like Google's Gemma and OpenAI's GPT offer a promising solution to the challenge of obtaining large, labeled datasets for training classifiers. When rapid model deployment is critical, such as in classifying emerging social media trends or combating new forms of online abuse tied to current events, the ability to generate training data is invaluable. While prior research has examined the comparability of synthetic data to human-labeled data, this study introduces a novel sampling algorithm, based on the maximum coverage problem, to select a representative subset from a synthetically generated dataset. Our results demonstrate that training a classifier on this contextually sampled subset achieves superior performance compared to training on the entire dataset. This "less is more" approach not only improves model accuracy but also reduces the volume of data required, leading to potentially more efficient model fine-tuning.

cs.LG↗

arXiv2024

The ParClusterers Benchmark Suite (PCBS): A Fine-Grained Analysis of Scalable Graph Clustering

We introduce the ParClusterers Benchmark Suite (PCBS) -- a collection of highly scalable parallel graph clustering algorithms and benchmarking tools that streamline comparing different graph clustering algorithms and implementations. The benchmark includes clustering algorithms that target a wide range of modern clustering use cases, including community detection, classification, and dense subgraph mining. The benchmark toolkit makes it easy to run and evaluate multiple instances of different clustering algorithms, which can be useful for fine-tuning the performance of clustering on a given task, and for comparing different clustering algorithms based on different metrics of interest, including clustering quality and running time. Using PCBS, we evaluate a broad collection of real-world graph clustering datasets. Somewhat surprisingly, we find that the best quality results are obtained by algorithms that not included in many popular graph clustering toolkits. The PCBS provides a standardized way to evaluate and judge the quality-performance tradeoffs of the active research area of scalable graph clustering algorithms. We believe it will help enable fair, accurate, and nuanced evaluation of graph clustering algorithms in the future.

cs.DC↗

arXiv2023

Two-Step Active Learning for Instance Segmentation with Uncertainty and Diversity Sampling

Training high-quality instance segmentation models requires an abundance of labeled images with instance masks and classifications, which is often expensive to procure. Active learning addresses this challenge by striving for optimum performance with minimal labeling cost by selecting the most informative and representative images for labeling. Despite its potential, active learning has been less explored in instance segmentation compared to other tasks like image classification, which require less labeling. In this study, we propose a post-hoc active learning algorithm that integrates uncertainty-based sampling with diversity-based sampling. Our proposed algorithm is not only simple and easy to implement, but it also delivers superior performance on various datasets. Its practical application is demonstrated on a real-world overhead imagery dataset, where it increases the labeling efficiency fivefold.

cs.CV↗

arXiv2023

Agile Modeling: From Concept to Classifier in Minutes

The application of computer vision to nuanced subjective use cases is growing. While crowdsourcing has served the vision community well for most objective tasks (such as labeling a "zebra"), it now falters on tasks where there is substantial subjectivity in the concept (such as identifying "gourmet tuna"). However, empowering any user to develop a classifier for their concept is technically difficult: users are neither machine learning experts, nor have the patience to label thousands of examples. In reaction, we introduce the problem of Agile Modeling: the process of turning any subjective visual concept into a computer vision model through a real-time user-in-the-loop interactions. We instantiate an Agile Modeling prototype for image classification and show through a user study (N=14) that users can create classifiers with minimal effort under 30 minutes. We compare this user driven process with the traditional crowdsourcing paradigm and find that the crowd's notion often differs from that of the user's, especially as the concepts become more subjective. Finally, we scale our experiments with simulations of users training classifiers for ImageNet21k categories to further demonstrate the efficacy.

cs.LG↗

arXiv2022

A First-Order Numerical Algorithm without Matrix Operations

This paper offers a matrix-free first-order numerical method to solve large-scale conic optimization problems. Solving systems of linear equations pose the most computationally challenging part in both first-order and second-order numerical algorithms. Existing direct and indirect methods are either computationally expensive or compromise on solution accuracy. Alternatively, we propose an easy-to-compute decomposition method to solve sparse linear systems that arise in conic optimization problems. Its iterations are tractable, highly parallelizable, with closed-form solutions. This algorithm can be easily implemented on distributed platforms, such as graphics processing units, with orders-of-magnitude time improvement. The performance of the proposed solver is demonstrated on large-scale conic optimization problems and is compared with the state-of-the-art first-order solvers.

math.OC↗

arXiv2021

An Automatic Approach for Generating Rich, Linked Geo-Metadata from Historical Map Images

Historical maps contain detailed geographic information difficult to find elsewhere covering long-periods of time (e.g., 125 years for the historical topographic maps in the US). However, these maps typically exist as scanned images without searchable metadata. Existing approaches making historical maps searchable rely on tedious manual work (including crowd-sourcing) to generate the metadata (e.g., geolocations and keywords). Optical character recognition (OCR) software could alleviate the required manual work, but the recognition results are individual words instead of location phrases (e.g., "Black" and "Mountain" vs. "Black Mountain"). This paper presents an end-to-end approach to address the real-world problem of finding and indexing historical map images. This approach automatically processes historical map images to extract their text content and generates a set of metadata that is linked to large external geospatial knowledge bases. The linked metadata in the RDF (Resource Description Framework) format support complex queries for finding and indexing historical maps, such as retrieving all historical maps covering mountain peaks higher than 1,000 meters in California. We have implemented the approach in a system called mapKurator. We have evaluated mapKurator using historical maps from several sources with various map styles, scales, and coverage. Our results show significant improvement over the state-of-the-art methods. The code has been made publicly available as modules of the Kartta Labs project at https://github.com/kartta-labs/Project.

cs.AI↗

arXiv2021

Rapid Convergence of First-Order Numerical Algorithms via Adaptive Conditioning

This paper is an attempt to remedy the problem of slow convergence for first-order numerical algorithms by proposing an adaptive conditioning heuristic. First, we propose a parallelizable numerical algorithm that is capable of solving large-scale conic optimization problems on distributed platforms such as {graphics processing unit} with orders-of-magnitude time improvement. Proof of global convergence is provided for the proposed algorithm. We argue that on the contrary to common belief, the condition number of the data matrix is not a reliable predictor of convergence speed. In light of this observation, an adaptive conditioning heuristic is proposed which enables higher accuracy compared to other first-order numerical algorithms. Numerical experiments on a wide range of large-scale linear programming and second-order cone programming problems demonstrate the scalability and computational advantages of the proposed algorithm compared to commercial and open-source state-of-the-art solvers.

math.OC↗

arXiv2020

Adaptive Third Order Adams-Bashforth Time Stepping for Extended Boussinesq Equations

We develop the third-order adaptive Adams-Bashforth time stepping and the second-order finite difference equation for variable time steps. We incorporate these schemes in the Celeris Advent software to discretize and solve the 2D extended Boussinesq equations. This software uses a hybrid finite volume - finite difference scheme and leverages the GPU to solve the equations faster than real-time while concurrently visualizing them. We simulate several benchmarks using the adaptive time stepping scheme of Celeris Advent and demonstrate the capability of the software in modeling wave-breaking, wave runup, irregular waves, and rip currents. The adaptive scheme significantly improves the robustness of the model while providing faster computational performance.

math.NA↗

arXiv2020

Kartta Labs: Collaborative Time Travel

We introduce the modular and scalable design of Kartta Labs, an open source, open data, and scalable system for virtually reconstructing cities from historical maps and photos. Kartta Labs relies on crowdsourcing and artificial intelligence consisting of two major modules: Maps and 3D models. Each module, in turn, consists of sub-modules that enable the system to reconstruct a city from historical maps and photos. The result is a spatiotemporal reference that can be used to integrate various collected data (curated, sensed, or crowdsourced) for research, education, and entertainment purposes. The system empowers the users to experience collaborative time travel such that they work together to reconstruct the past and experience it on an open source and open data platform.

cs.CV↗

arXiv2017

A Review on the Applications of Crowdsourcing in Human Pathology

The advent of the digital pathology has introduced new avenues of diagnostic medicine. Among them, crowdsourcing has attracted researchers' attention in the recent years, allowing them to engage thousands of untrained individuals in research and diagnosis. While there exist several articles in this regard, prior works have not collectively documented them. We, therefore, aim to review the applications of crowdsourcing in human pathology in a semi-systematic manner. We firstly, introduce a novel method to do a systematic search of the literature. Utilizing this method, we, then, collect hundreds of articles and screen them against a pre-defined set of criteria. Furthermore, we crowdsource part of the screening process, to examine another potential application of crowdsourcing. Finally, we review the selected articles and characterize the prior uses of crowdsourcing in pathology.

cs.HC↗

arXiv2017

Celeris: A GPU-accelerated open source software with a Boussinesq-type wave solver for real-time interactive simulation and visualization

In this paper, we introduce an interactive coastal wave simulation and visualization software, called Celeris. Celeris is an open source software which needs minimum preparation to run on a Windows machine. The software solves the extended Boussinesq equations using a hybrid finite volume - finite difference method and supports moving shoreline boundaries. The simulation and visualization are performed on the GPU using Direct3D libraries, which enables the software to run faster than real-time. Celeris provides a first-of-its-kind interactive modeling platform for coastal wave applications and it supports simultaneous visualization with both photorealistic and colormapped rendering capabilities. We validate our software through comparison with three standard benchmarks for non-breaking and breaking waves.

physics.flu-dyn↗

arXiv2017

Wall and Bed Shear Force in Rectangular Open Channels

This paper studies the effects of velocity gradients and secondary currents on the distribution of the shear force between the walls and bed of rectangular open channels. We show that neglecting the effect of secondary currents and assuming zero-shear division lines does not yield acceptable results. We, accordingly, introduce a method to determine the percentage of the total shear force acting on the walls and bed of rectangular open channels, which takes both the velocity gradients and secondary currents into account. Using the channel bisectors, along which there is no secondary flows effect, and orthogonal trajectories to isovels, along which there is no shear stress, we divide the channel cross section into three major subsections, namely bed area, walls area, and shared area. The geometry of each subsection is derived given the location of the maximum velocity. The share of the bed and wall shear forces from the shared area, are calculated afterward. The results for bed and walls shear forces agree well with the experimental data with an average relative error less than 5% for both regular flows and flows carrying suspended sediment. This method also provides a physics-driven range for the wall and bed shear forces which nicely covers the experimental data.

physics.flu-dyn↗