Artificial Intelligence and Robotics Commons^™

Transformers In Remote Sensing: A Survey, Abdulaziz Amer Aleissaee, Amandeep Kumar, Rao Anwer, Salman Khan, Hisham Cholakkal, Gui-Song Xia, Fahad Shahbaz Khan

Computer Vision Faculty Publications

Deep learning-based algorithms have seen a massive popularity in different areas of remote sensing image analysis over the past decade. Recently, transformers-based architectures, originally introduced in natural language processing, have pervaded computer vision field where the self-attention mechanism has been utilized as a replacement to the popular convolution operator for capturing long-range dependencies. Inspired by recent advances in computer vision, remote sensing community has also witnessed an increased exploration of vision transformers for a diverse set of tasks. Although a number of surveys have focused on transformers in computer vision in general, to the best of our knowledge we are …

Accomontage2: A Complete Harmonization And Accompaniment Arrangement System, Li Yi, Haochen Hu, Jingwei Zhao, Gus Xia

Machine Learning Faculty Publications

We propose AccoMontage2, a system capable of doing full-length song harmonization and accompaniment arrangement based on a lead melody. Following AccoMontage, this study focuses on generating piano arrangements for popular/folk songs and it carries on the generalized template-based retrieval method. The novelties of this study are twofold. First, we invent a harmonization module (which AccoMontage does not have). This module generates structured and coherent full-length chord progression by optimizing and balancing three loss terms: a micro-level loss for note-wise dissonance, a meso-level loss for phrase-template matching, and a macro-level loss for full piece coherency. Second, we develop a graphical user …

Overview Of The Clef-2022 Checkthat! Lab Task 2 On Detecting Previously Fact-Checked Claims, Preslav Nakov, Giovanni Da San Martino, Firoj Alam, Shaden Shaar, Hamdy Mubarak, Nikolay Babulkov

Natural Language Processing Faculty Publications

We describe the fourth edition of the CheckThat! Lab, part of the 2022 Conference and Labs of the Evaluation Forum (CLEF). The lab evaluates technology supporting three tasks related to factuality, and it covers seven languages such as Arabic, Bulgarian, Dutch, English, German, Spanish, and Turkish. Here, we present the task 2, which asks to detect previously fact-checked claims (in two languages). A total of six teams participated in this task, submitted a total of 37 runs, and most submissions managed to achieve sizable improvements over the baselines using transformer based models such as BERT, RoBERTa. In this paper, we …

Negational Symmetry Of Quantum Neural Networks For Binary Pattern Classification, Nanqing Dong, Michael Kampffmeyer, Irina Voiculescu, Eric P. Xing

Machine Learning Faculty Publications

Although quantum neural networks (QNNs) have shown promising results in solving simple machine learning tasks recently, the behavior of QNNs in binary pattern classification is still underexplored. In this work, we find that QNNs have an Achilles’ heel in binary pattern classification. To illustrate this point, we provide a theoretical insight into the properties of QNNs by presenting and analyzing a new form of symmetry embedded in a family of QNNs with full entanglement, which we term negational symmetry. Due to negational symmetry, QNNs can not differentiate between a quantum binary signal and its negational counterpart. We empirically evaluate the …

Overview Of The Clef-2022 Checkthat! Lab Task 1 On Identifying Relevant Claims In Tweets, Preslav Nakov, Alberto Barrón-Cedeño, Giovanni Da San Martino, Firoj Alam, Mucahid Kutlu, Wajdi Zaghouani, Mucahid Kutlu, Wajdi Zaghouani, Chengkai Li, Shaden Shaar, Hamdy Mubarak, Alex Nikolov

Natural Language Processing Faculty Publications

We present an overview of CheckThat! lab 2022 Task 1, part of the 2022 Conference and Labs of the Evaluation Forum (CLEF). Task 1 asked to predict which posts in a Twitter stream are worth fact-checking, focusing on COVID-19 and politics in six languages: Arabic, Bulgarian, Dutch, English, Spanish, and Turkish. A total of 19 teams participated and most submissions managed to achieve sizable improvements over the baselines using Transformer-based models such as BERT and GPT-3. Across the four subtasks, approaches that targetted multiple languages (be it individually or in conjunction, in general obtained the best performance. We describe the …

Artificial Intelligence And Human Employment, Singapore Management University

Perspectives@SMU

AI will replace humans in repetitive tasks. Greater value can be created when it augments and complements the jobs people do

Singapore Public Sector Ai Applications Emphasizing Public Engagement: Six Examples, Steven M. Miller

Research Collection School Of Computing and Information Systems

This article provides an overview of six examples of public sector AI applications in Singapore that illustrate different ways of enhancing engagement with the public. These applications demonstrate ways of enhancing engagement with the public by providing greater accessibility to government services (access anywhere, anytime) and speedier responses to public processes and feedback. Some applications make it substantially easier for members of the public to do things or make choices, while others reduce waiting time, either across an entire public infrastructure, or for an individual transaction. Some provide highly individualized coaching to guide a person through the process of doing …

Products Pricing And Return Strategies For The Dual Channel Retailers, Jian Liu, Xinyue Sun, Yanyan Liu

Electrical and Computer Engineering Faculty Research & Creative Works

This paper analyzed how different return strategies and return rates affect dual-channel retailers' profits and channel pricings. Return can stimulate sales; however, the return has presented significant challenges to retailers. The return has long been studied to maximize profit and pricing; however, the different return strategies for dual-channel retailers affect channel both. This paper aimed to study whether or not dual-channel retailers should allow customers to return items in two channels and whether or not the retailer should contract with the manufacturers and pay extra fees to return products. This study indicated when the retailer should allow customers' returns to …

Leveraging Subject Matter Expertise To Optimize Machine Learning Techniques For Air And Space Applications, Philip Y. Cho

Theses and Dissertations

We develop new machine learning and statistical methods that are tailored for Air and Space applications through the incorporation of subject matter expertise. In particular, we focus on three separate research thrusts that each represents a different type of subject matter knowledge, modeling approach, and application. In our first thrust, we incorporate knowledge of natural phenomena to design a neural network algorithm for localizing point defects in transmission electron microscopy (TEM) images of crystalline materials. In our second research thrust, we use Bayesian feature selection and regression to analyze the relationship between fighter pilot attributes and flight mishap rates. We …

Degrees Of Confidence As A Legal Tool To Assess Ai System Liability, Joshua Song

Michigan Technology Law Review

AI systems have become increasingly integrated into our everyday lives, and harms caused by these systems have graduated from raising hypothetical ethical concerns to questions of actual legal liability. Civil liability schemes are generally designed to address harms caused by humans; thus, it may be tempting to analogize new types of harms caused by AI systems to familiar harms caused by humans in order to justify commandeering existing human-centered legal tools to assess AI liability. However, the analogy is inappropriate and misrepresents salient legal differences in how harms are committed by humans and AI systems. Thus, “as is often the …

Influence Level Prediction On Social Media Through Multi-Task And Sociolinguistic User Characteristics Modeling, Denys Katerenchuk

Dissertations, Theses, and Capstone Projects

Prediction of a user’s influence level on social networks has attracted a lot of attention as human interactions move online. Influential users have the ability to influence others’ behavior to achieve their own agenda. As a result, predicting users’ level of influence online can help to understand social networks, forecast trends, prevent misinformation, etc. The research on user influence in social networks has attracted much attention across multiple disciplines, from social sciences to mathematics, yet it is still not well understood. One of the difficulties is that the definition of influence is specific to a particular problem or a domain, …

Data-Centric Machine Learning For Speech And Audio, Ali Raza Syed

Dissertations, Theses, and Capstone Projects

There is growing recognition of the importance of data-centric methods for building machine learning systems. Data-centric methods assume a fixed model and iterate over the data to improve system performance. This is in contrast to traditional model-centric approaches, which assume a fixed dataset and iterate over models for the same ends. Data-centric machine learning is driven by the observation that, beyond the size of the training data, model performance depends on factors such as the quality of the annotations, and whether the data are representative of conditions in which models will be deployed. This is particularly of interest in the …

The Interaction Of Different Primary Producers And Physical And Chemical Dynamics Of An Urban Shallow Lake, Majid Sahin

Dissertations, Theses, and Capstone Projects

An artificial urban shallow lake, Prospect Park Lake (PPL), is situated on a terminal moraine in Brooklyn New York, and supplied with municipal water treated with ortho-phosphates. The constant input of the phosphate nutrient is the primary source of eutrophication in the lake. The numerous pools along the water course houses various aquatic phototrophs, which influence the water quality and the state of the system, driving conditions into favoring the survival of their species. In the first half of the dissertation, the focus of the project is on analyzing how the different primary producers in different regions of PPL affect …

Finite Gaussian Neurons: Defending Against Adversarial Attacks By Making Neural Networks Say "I Don’T Know", Felix Grezes

Dissertations, Theses, and Capstone Projects

In this work, I introduce the Finite Gaussian Neuron (FGN), a novel neuron architecture for artificial neural networks aimed at protecting against adversarial attacks.
Since 2014, artificial neural networks have been known to be vulnerable to adversarial attacks, which can fool the network into producing wrong or nonsensical outputs by making humanly imperceptible alterations to inputs. While defenses against adversarial attacks have been proposed, they usually involve retraining a new neural network from scratch, a costly task.

My works aims to:
- easily convert existing models to Finite Gaussian Neuron architecture,
- while preserving the existing model's behavior on real …

A Carbon-Aware Planning Framework For Production Scheduling In Mining, Nurual Asyikeen Azhar, Aldy Gunawan, Shih-Fen Cheng, Erwin Leonardi

Research Collection School Of Computing and Information Systems

Managing the flow of excavated materials from a mine pit and the subsequent processing steps is the logistical challenge in mining. Mine planning needs to consider various geometric and resource constraints while maximizing the net present value (NPV) of profits over a long horizon. This mine planning problem has been modelled and solved as a precedence constrained production scheduling problem (PCPSP) using heuristics, due to its NP-hardness. However, the recent push for sustainable and carbon-aware mining practices calls for new planning approaches. In this paper, we propose an efficient temporally decomposed greedy Lagrangian relaxation (TDGLR) approach to maximize profits while …

Contrastive Transformer-Based Multiple Instance Learning For Weakly Supervised Polyp Frame Detection, Tian Yu, Guansong Pang, Fengbei Liu, Yuyuan Liu, Chong Wang, Yuanhong Chen, Johan Verjans, Gustavo Carneiro

Research Collection School Of Computing and Information Systems

Current polyp detection methods from colonoscopy videos use exclusively normal (i.e., healthy) training images, which i) ignore the importance of temporal information in consecutive video frames, and ii) lack knowledge about the polyps. Consequently, they often have high detection errors, especially on challenging polyp cases (e.g., small, flat, or partially visible polyps). In this work, we formulate polyp detection as a weakly-supervised anomaly detection task that uses video-level labelled training data to detect frame-level polyps. In particular, we propose a novel convolutional transformer-based multiple instance learning method designed to identify abnormal frames (i.e., frames with polyps) from anomalous videos (i.e., …

Deep Learning-Based Text Recognition Of Agricultural Regulatory Document, Hua Leong Fwa, Farn Haur Chan

Research Collection School Of Computing and Information Systems

In this study, an OCR system based on deep learning techniques was deployed to digitize scanned agricultural regulatory documents comprising of certificates and labels. Recognition of the certificates and labels is challenging as they are scanned images of the hard copy form and the layout and size of the text as well as the languages vary between the various countries (due to diverse regulatory requirements). We evaluated and compared between various state-of-the-art deep learningbased text detection and recognition model as well as a packaged OCR library – Tesseract. We then adopted a two-stage approach comprising of text detection using Character …

Deep Learning For Coverage-Guided Fuzzing: How Far Are We?, Siqi Li, Xiaofei Xie, Yun Lin, Yuekang Li, Ruitao Feng, Xiaohong Li, Weimin Ge, Jin Song Dong

Research Collection School Of Computing and Information Systems

Fuzzing is a widely-used software vulnerability discovery technology, many of which are optimized using coverage-feedback. Recently, some techniques propose to train deep learning (DL) models to predict the branch coverage of an arbitrary input owing to its always-available gradients etc. as a guide. Those techniques have proved their success in improving coverage and discovering bugs under different experimental settings. However, DL models, usually as a magic black-box, are notoriously lack of explanation. Moreover, their performance can be sensitive to the collected runtime coverage information for training, indicating potentially unstable performance. In this work, we conduct a systematic empirical study on …

Two-Phase Matheuristic For The Vehicle Routing Problem With Reverse Cross-Docking, Aldy Gunawan, Audrey Tedja Widjaja, Pieter Vansteenwegen, Vincent F. Yu

Research Collection School Of Computing and Information Systems

Cross-dockingis a useful concept used by many companies to control the product flow. It enables the transshipment process of products from suppliers to customers. This research thus extends the benefit of cross-docking with reverse logistics, since return process management has become an important field in various businesses. The vehicle routing problem in a distribution network is considered to be an integrated model, namely the vehicle routing problem with reverse cross-docking (VRP-RCD). This study develops a mathematical model to minimize the costs of moving products in a four-level supply chain network that involves suppliers, cross-dock, customers, and outlets. A matheuristic based …

Constrained Multiagent Reinforcement Learning For Large Agent Population, Jiajing Ling, Arambam James Singh, Duc Thien Nguyen, Akshat Kumar

Research Collection School Of Computing and Information Systems

Learning control policies for a large number of agents in a decentralized setting is challenging due to partial observability, uncertainty in the environment, and scalability challenges. While several scalable multiagent RL (MARL) methods have been proposed, relatively few approaches exist for large scale constrained MARL settings. To address this, we first formulate the constrained MARL problem in a collective multiagent setting where interactions among agents are governed by the aggregate count and types of agents, and do not depend on agents’ specific identities. Second, we show that standard Lagrangian relaxation methods, which are popular for single agent RL, do not …