arXiv subjects

Milton Halem

Publications and source records attributed to Milton Halem.

12 recordsLinked to original sources

arXiv2026

WildfireVLM: AI-powered Analysis for Early Wildfire Detection and Risk Assessment Using Satellite Imagery

Wildfires are a growing threat to ecosystems, human lives, and infrastructure, with their frequency and intensity rising due to climate change and human activities. Early detection is critical, yet satellite-based monitoring remains challenging due to faint smoke signals, dynamic weather conditions, and the need for real-time analysis over large areas. We introduce WildfireVLM, an AI framework that combines satellite imagery wildfire detection with language-driven risk assessment. We construct a labeled wildfire and smoke dataset using imagery from Landsat-8/9, GOES-16, and other publicly available Earth observation sources, including harmonized products with aligned spectral bands. WildfireVLM employs YOLOv12 to detect fire zones and smoke plumes, leveraging its ability to detect small, complex patterns in satellite imagery. We integrate Multimodal Large Language Models (MLLMs) that convert detection outputs into contextualized risk assessments and prioritized response recommendations for disaster management. We validate the quality of risk reasoning using an LLM-as-judge evaluation with a shared rubric. The system is deployed using a service-oriented architecture that supports real-time processing, visual risk dashboards, and long-term wildfire tracking, demonstrating the value of combining computer vision with language-based reasoning for scalable wildfire monitoring. The code and dataset are publicly available on GitHub at https://github.com/Ayanzadeh93/_WildfireVLM_.

cs.CV↗

arXiv2024

DUNE: A Machine Learning Deep UNet++ based Ensemble Approach to Monthly, Seasonal and Annual Climate Forecasting

Capitalizing on the recent availability of ERA5 monthly averaged long-term data records of mean atmospheric and climate fields based on high-resolution reanalysis, deep-learning architectures offer an alternative to physics-based daily numerical weather predictions for subseasonal to seasonal (S2S) and annual means. A novel Deep UNet++-based Ensemble (DUNE) neural architecture is introduced, employing multi-encoder-decoder structures with residual blocks. When initialized from a prior month or year, this architecture produced the first AI-based global monthly, seasonal, or annual mean forecast of 2-meter temperatures (T2m) and sea surface temperatures (SST). ERA5 monthly mean data is used as input for T2m over land, SST over oceans, and solar radiation at the top of the atmosphere for each month of 40 years to train the model. Validation forecasts are performed for an additional two years, followed by five years of forecast evaluations to account for natural annual variability. AI-trained inference forecast weights generate forecasts in seconds, enabling ensemble seasonal forecasts. Root Mean Squared Error (RMSE), Anomaly Correlation Coefficient (ACC), and Heidke Skill Score (HSS) statistics are presented globally and over specific regions. These forecasts outperform persistence, climatology, and multiple linear regression for all domains. DUNE forecasts demonstrate comparable statistical accuracy to NOAA's operational monthly and seasonal probabilistic outlook forecasts over the US but at significantly higher resolutions. RMSE and ACC error statistics for other recent AI-based daily forecasts also show superior performance for DUNE-based forecasts. The DUNE model's application to an ensemble data assimilation cycle shows comparable forecast accuracy with a single high-resolution model, potentially eliminating the need for retraining on extrapolated datasets.

cs.LG↗

arXiv2022

Quantum-Assisted Greedy Algorithms

We show how to leverage quantum annealers (QAs) to better select candidates in greedy algorithms. Unlike conventional greedy algorithms that employ problem-specific heuristics for making locally optimal choices at each stage, we use QAs that sample from the ground state of problem-dependent Hamiltonians at cryogenic temperatures and use retrieved samples to estimate the probability distribution of problem variables. More specifically, we look at each spin of the Ising model as a random variable and contract all problem variables whose corresponding uncertainties are negligible. Our empirical results on a D-Wave 2000Q quantum processor demonstrate that the proposed quantum-assisted greedy algorithm (QAGA) scheme can find notably better solutions compared to the state-of-the-art techniques in the realm of quantum annealing

quant-ph↗

arXiv2020

Multi-Qubit Correction for Quantum Annealers

We present \emph{multi-qubit correction} (MQC) as a novel postprocessing method for quantum annealers that views the evolution in an open-system as a Gibbs sampler and reduces a set of excited states to a new synthetic state with lower energy value. After sampling from the ground state of a given (Ising) Hamiltonian, MQC compares pairs of excited states to recognize virtual tunnels--i.e., a group of qubits that changing their states simultaneously can result in a new state with lower energy value--and successively converges to the ground state. Experimental results using D-Wave 2000Q quantum annealers demonstrate that MQC finds samples with notably lower energy values and improves the reproducibility of results when compared to recent hardware/software advances in the realm of quantum annealing, such as spin-reversal transforms, classical postprocessing techniques, and increased inter-sample delay between successive measurements.

quant-ph↗

arXiv2020

An Ensemble Approach for Compressive Sensing with Quantum

We leverage the idea of a statistical ensemble to improve the quality of quantum annealing based binary compressive sensing. Since executing quantum machine instructions on a quantum annealer can result in an excited state, rather than the ground state of the given Hamiltonian, we use different penalty parameters to generate multiple distinct quadratic unconstrained binary optimization (QUBO) functions whose ground state(s) represent a potential solution of the original problem. We then employ the attained samples from minimizing all corresponding (different) QUBOs to estimate the solution of the problem of binary compressive sensing. Our experiments, on a D-Wave 2000Q quantum processor, demonstrated that the proposed ensemble scheme is notably less sensitive to the calibration of the penalty parameter that controls the trade-off between the feasibility and sparsity of recoveries.

quant-ph↗

arXiv2020

A Hybrid Quantum enabled RBM Advantage: Convolutional Autoencoders For Quantum Image Compression and Generative Learning

Understanding how the D-Wave quantum computer could be used for machine learning problems is of growing interest. Our work evaluates the feasibility of using the D-Wave as a sampler for machine learning. We describe a hybrid system that combines a classical deep neural network autoencoder with a quantum annealing Restricted Boltzmann Machine (RBM) using the D-Wave. We evaluate our hybrid autoencoder algorithm using two datasets, the MNIST dataset and MNIST Fashion dataset. We evaluate the quality of this method by using a downstream classification method where the training is based on quantum RBM-generated samples. Our method overcomes two key limitations in the current 2000-qubit D-Wave processor, namely the limited number of qubits available to accommodate typical problem sizes for fully connected quantum objective functions and samples that are binary pixel representations. As a consequence of these limitations we are able to show how we achieved nearly a 22-fold compression factor of grayscale 28 x 28 sized images to binary 6 x 6 sized images with a lossy recovery of the original 28 x 28 grayscale images. We further show how generating samples from the D-Wave after training the RBM, resulted in 28 x 28 images that were variations of the original input data distribution, as opposed to recreating the training samples. We formulated an MNIST classification problem using a deep convolutional neural network that used samples from a quantum RBM to train the MNIST classifier and compared the results with an MNIST classifier trained with the original MNIST training data set, as well as an MNIST classifier trained using classical RBM samples. Our hybrid autoencoder approach indicates advantage for RBM results relative to the use of a current RBM classical computer implementation for image-based machine learning and even more promising results for the next generation D-Wave quantum system.

quant-ph↗

arXiv2020

Reinforcement Quantum Annealing: A Quantum-Assisted Learning Automata Approach

We introduce the reinforcement quantum annealing (RQA) scheme in which an intelligent agent interacts with a quantum annealer that plays the stochastic environment role of learning automata and tries to iteratively find better Ising Hamiltonians for the given problem of interest. As a proof-of-concept, we propose a novel approach for reducing the NP-complete problem of Boolean satisfiability (SAT) to minimizing Ising Hamiltonians and show how to apply the RQA for increasing the probability of finding the global optimum. Our experimental results on two different benchmark SAT problems (namely factoring pseudo-prime numbers and random SAT with phase transitions), using a D-Wave 2000Q quantum processor, demonstrated that RQA finds notably better solutions with fewer samples, compared to state-of-the-art techniques in the realm of quantum annealing.

quant-ph↗

arXiv2019

Quantum-Assisted Greedy Algorithms

We show how to leverage quantum annealers to better select candidates in greedy algorithms. Unlike conventional greedy algorithms that employ problem-specific heuristics for making locally optimal choices at each stage, we use quantum annealers that sample from the ground state(s) of a problem-dependent Ising Hamiltonians at cryogenic temperatures and use retrieved samples to estimate the probability distribution of problem variables. More specifically, we look at each spin of the Ising model as a random variable and contract all problem variables whose corresponding uncertainties are negligible. Our empirical results, on a D-Wave 2000Q quantum processor, revealed that the proposed quantum-assisted greedy algorithm (QAGA) can find notably better solutions (i.e., samples with lower energy value), compared to the state-of-the-art techniques in the realm of quantum annealing.

quant-ph↗

arXiv2019

Exploring the Behavior of Coherent Accelerator Processor Interface (CAPI) on IBM Power8+ Architecture and FlashSystem 900

The Coherent Accelerator Processor Interface (CAPI) is a general term for the infrastructure that provides high throughput and low latency path to the flash storage connected to the IBM POWER 8+ System. CAPI accelerator card is attached coherently as a peer to the Power8+ processor. This removes the overhead and complexity of the IO subsystem and allows the accelerator to operate as part of an application. In this paper, we present the results of experiments on IBM FlashSystem900 (FS900) with CAPI accelerator card using the "CAPI-Flash IBM Data Engine for NoSQL Software" Library. This library provides the application, a direct access to the underlying flash storage through user space APIs, to manage and access the data in flash. This offloads kernel IO driver functionality to dedicated CAPI FPGA accelerator hardware. We conducted experiments to analyze the performance of FS900 with CAPI accelerator card, using the Key Value Layer APIs, employing NASA's MODIS Land Surface Reflectance dataset as a large dataset use case. We performed Read and Write operations on datasets of size ranging from 1MB to 3TB by varying the number of threads. We then compared this performance with other heterogeneous storage and memory devices such as NVM, SSD and RAM, without using the CAPI Accelerator in synchronous and asynchronous file IO modes of operations. The results indicate that FS900 & CAPI, together with the metadata cache in RAM, delivers the highest IO/s and OP/s for read operations. This was higher than just using RAM, along with utilizing lesser CPU resources. Among FS900, SSD and NVM, FS900 had the highest write IO/s. Another important observation is that, when the size of the input dataset exceeds the capacity of RAM, and when the data access is non-uniform and sparse, FS900 with CAPI would be a cost-effective alternative.

cs.DC↗

arXiv2019

SAT-based Compressive Sensing

We propose to reduce the original well-posed problem of compressive sensing to weighted-MAX-SAT. Compressive sensing is a novel randomized data acquisition approach that linearly samples sparse or compressible signals at a rate much below the Nyquist-Shannon sampling rate. The original problem of compressive sensing in sparse recovery is NP-hard; therefore, in addition to restrictions for the uniqueness of the sparse solution, the coding matrix has also to satisfy additional stringent constraints -usually the restricted isometry property (RIP)- so we can handle it by its convex or nonconvex relaxations. In practice, such constraints are not only intractable to be verified but also invalid in broad applications. We first divide the well-posed problem of compressive sensing into relaxed sub-problems and represent them as separate SAT instances in conjunctive normal form (CNF). After merging the resulting sub-problems, we assign weights to all clauses in such a way that the aggregated weighted-MAX-SAT can guarantee successful recovery of the original signal. The only requirement in our approach is the solution uniqueness of the associated problems, which is notably looser. As a proof of concept, we demonstrate the applicability of our approach in tackling the original problem of binary compressive sensing with binary design matrices. Experimental results demonstrate the supremacy of the proposed SAT-based compressive sensing over the $\ell_1$-minimization in the robust recovery of sparse binary signals. SAT-based compressive sensing on average requires 8.3% fewer measurements for exact recovery of highly sparse binary signals ($s/N\approx 0.1$). When $s/N \approx 0.5$, the $\ell_1$-minimization on average requires 22.2% more measurements for exact reconstruction of the binary signals. Thus, the proposed SAT-based compressive sensing is less sensitive to the sparsity of the original signals.

cs.IT↗

arXiv2019

Quantum Annealing Based Binary Compressive Sensing with Matrix Uncertainty

Compressive sensing is a novel approach that linearly samples sparse or compressible signals at a rate much below the Nyquist-Shannon sampling rate and outperforms traditional signal processing techniques in acquiring and reconstructing such signals. Compressive sensing with matrix uncertainty is an extension of the standard compressive sensing problem that appears in various applications including but not limited to cognitive radio sensing, calibration of the antenna, and deconvolution. The original problem of compressive sensing is NP-hard so the traditional techniques, such as convex and nonconvex relaxations and greedy algorithms, apply stringent constraints on the measurement matrix to indirectly handle this problem in the realm of classical computing. We propose well-posed approaches for both binary compressive sensing and binary compressive sensing with matrix uncertainty problems that are tractable by quantum annealers. Our approach formulates an Ising model whose ground state represents a sparse solution for the binary compressive sensing problem and then employs an alternating minimization scheme to tackle the binary compressive sensing with matrix uncertainty problem. This setting only requires the solution uniqueness of the considered problem to have a successful recovery process, and therefore the required conditions on the measurement matrix are notably looser. As a proof of concept, we can demonstrate the applicability of the proposed approach on the D-Wave quantum annealers; however, we can adapt our method to employ other modern computing phenomena -like adiabatic quantum computers (in general), CMOS annealers, optical parametric oscillators, and neuromorphic computing.

cs.IT↗

arXiv2018

Ontology-Grounded Topic Modeling for Climate Science Research

In scientific disciplines where research findings have a strong impact on society, reducing the amount of time it takes to understand, synthesize and exploit the research is invaluable. Topic modeling is an effective technique for summarizing a collection of documents to find the main themes among them and to classify other documents that have a similar mixture of co-occurring words. We show how grounding a topic model with an ontology, extracted from a glossary of important domain phrases, improves the topics generated and makes them easier to understand. We apply and evaluate this method to the climate science domain. The result improves the topics generated and supports faster research understanding, discovery of social networks among researchers, and automatic ontology generation.

cs.CL↗