Artificial Intelligence and Robotics Commons^™

Conditional Neural Heuristic For Multiobjective Vehicle Routing Problems, Mingfeng Fan, Yaoxin Wu, Zhiguang Cao, Wen Song, Guillaume Sartoretti, Huan Liu, Guohua Wu

Research Collection School Of Computing and Information Systems

Existing neural heuristics for multiobjective vehicle routing problems (MOVRPs) are primarily conditioned on instance context, which failed to appropriately exploit preference and problem size, thus holding back the performance. To thoroughly unleash the potential, we propose a novel conditional neural heuristic (CNH) that fully leverages the instance context, preference, and size with an encoder–decoder structured policy network. Particularly, in our CNH, we design a dual-attention-based encoder to relate preferences and instance contexts, so as to better capture their joint effect on approximating the exact Pareto front (PF). We also design a size-aware decoder based on the sinusoidal encoding to explicitly …

A Chinese Power Text Classification Algorithm Based On Deep Active Learning, Song Deng, Qianliang Li, Renjie Dai, Siming Wei, Di Wu, Yi He, Xindong Wu

Computer Science Faculty Publications

The construction of knowledge graph is beneficial for grid production, electrical safety protection, fault diagnosis and traceability in an observable and controllable way. Highly-precision text classification algorithm is crucial to build a professional knowledge graph in power system. Unfortunately, there are a large number of poorly described and specialized texts in the power business system, and the amount of data containing valid labels in these texts is low. This will bring great challenges to improve the precision of text classification models. To offset the gap, we propose a classification algorithm for Chinese text in the power system based on deep …

Learning Optimal Inter-Class Margin Adaptively For Few-Shot Class-Incremental Learning Via Neural Collapse-Based Meta-Learning, Hang Ran, Weijun Li, Lusi Li, Songsong Tian, Xin Ning, Prayag Tiwari

Computer Science Faculty Publications

Few-Shot Class-Incremental Learning (FSCIL) aims to learn new classes incrementally with a limited number of samples per class. It faces issues of forgetting previously learned classes and overfitting on few-shot classes. An efficient strategy is to learn features that are discriminative in both base and incremental sessions. Current methods improve discriminability by manually designing inter-class margins based on empirical observations, which can be suboptimal. The emerging Neural Collapse (NC) theory provides a theoretically optimal inter-class margin for classification, serving as a basis for adaptively computing the margin. Yet, it is designed for closed, balanced data, not for sequential or few-shot …

Accelerating Markov Chain Monte Carlo Sampling With Diffusion Models, N. T. Hunt-Smith, W. Melnitchouk, F. Ringer, N. Sato, A. W. Thomas, M. J. White

Physics Faculty Publications

Global fits of physics models require efficient methods for exploring high-dimensional and/or multimodal posterior functions. We introduce a novel method for accelerating Markov Chain Monte Carlo (MCMC) sampling by pairing a Metropolis-Hastings algorithm with a diffusion model that can draw global samples with the aim of approximating the posterior. We briefly review diffusion models in the context of image synthesis before providing a streamlined diffusion model tailored towards low-dimensional data arrays. We then present our adapted Metropolis-Hastings algorithm which combines local proposals with global proposals taken from a diffusion model that is regularly trained on the samples produced during the …

Data Science In Finance: Challenges And Opportunities, Xianrong Zheng, Elizabeth Gildea, Sheng Chai, Tongxiao Zhang, Shuxi Wang

Information Technology & Decision Sciences Faculty Publications

Data science has become increasingly popular due to emerging technologies, including generative AI, big data, deep learning, etc. It can provide insights from data that are hard to determine from a human perspective. Data science in finance helps to provide more personal and safer experiences for customers and develop cutting-edge solutions for a company. This paper surveys the challenges and opportunities in applying data science to finance. It provides a state-of-the-art review of financial technologies, algorithmic trading, and fraud detection. Also, the paper identifies two research topics. One is how to use generative AI in algorithmic trading. The other is …

Continual Learning, Fast And Slow, Quang Anh Pham, Chenghao Liu, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

According to the Complementary Learning Systems (CLS) theory (McClelland et al. 1995) in neuroscience, humans do effective continual learning through two complementary systems: a fast learning system centered on the hippocampus for rapid learning of the specifics, individual experiences; and a slow learning system located in the neocortex for the gradual acquisition of structured knowledge about the environment. Motivated by this theory, we propose DualNets (for Dual Networks), a general continual learning framework comprising a fast learning system for supervised learning of pattern-separated representation from specific tasks and a slow learning system for representation learning of task-agnostic general representation via …

Active Discovering New Slots For Task-Oriented Conversation, Yuxia Wu, Tianhao Dai, Zhedong Zheng, Lizi Liao

Research Collection School Of Computing and Information Systems

Existing task-oriented conversational systems heavily rely on domain ontologies with pre-defined slots and candidate values. In practical settings, these prerequisites are hard to meet, due to the emerging new user requirements and ever-changing scenarios. To mitigate these issues for better interaction performance, there are efforts working towards detecting out-of-vocabulary values or discovering new slots under unsupervised or semi-supervised learning paradigms. However, overemphasizing on the conversation data patterns alone induces these methods to yield noisy and arbitrary slot results. To facilitate the pragmatic utility, real-world systems tend to provide a stringent amount of human labeling quota, which offers an authoritative way …

Deep Learning Uncertainty Quantification For Clinical Text Classification, Alina Peluso, Ioana Danciu, Hong-Jun Yoon, Jamaludin Mohd Yusof, Tanmoy Bhattacharya, Adam Spannaus, Noah Schaefferkoetter, Eric B. Durbin, Xiao-Cheng Wu, Antoinette Stroup, Jennifer Doherty, Stephen Schwartz, Charles Wiggins, Linda Coyle, Lynne Penberthy, Georgia D. Tourassi, Shang Gao

School of Public Health Faculty Publications

INTRODUCTION: Machine learning algorithms are expected to work side-by-side with humans in decision-making pipelines. Thus, the ability of classifiers to make reliable decisions is of paramount importance. Deep neural networks (DNNs) represent the state-of-the-art models to address real-world classification. Although the strength of activation in DNNs is often correlated with the network's confidence, in-depth analyses are needed to establish whether they are well calibrated. METHOD: In this paper, we demonstrate the use of DNN-based classification tools to benefit cancer registries by automating information extraction of disease at diagnosis and at surgery from electronic text pathology reports from the US National …

Learning Program Semantics For Vulnerability Detection Via Vulnerability-Specific Inter-Procedural Slicing, Bozhi Wu, Shangqing Liu, Xiao Yang, Zhiming Li, Jun Sun, Shang-Wei Lin

Research Collection School Of Computing and Information Systems

Learning-based approaches that learn code representations for software vulnerability detection have been proven to produce inspiring results. However, they still fail to capture complete and precise vulnerability semantics for code representations. To address the limitations, in this work, we propose a learning-based approach namely SnapVuln, which first utilizes multiple vulnerability-specific inter-procedural slicing algorithms to capture vulnerability semantics of various types and then employs a Gated Graph Neural Network (GGNN) with an attention mechanism to learn vulnerability semantics. We compare SnapVuln with state-of-the-art learning-based approaches on two public datasets, and confirm that SnapVuln outperforms them. We further perform an ablation study …

Artificial Intelligence History, And Libraries: History And Legacy Of Library Contributions To Machine Learning, Wilhelmina Randtke

Library Faculty Presentations

Machine learning seems to be newly everywhere. It's not new, so much as faster processing makes it newly useful. Imagine an automated cataloging program that takes 300 years to run, versus one that takes a week to run. Increased processing speed is a substantive change. This presentation overviews the history of libraries and artificial intelligence. First, teasing out past applications of machine learning in libraries. High quality results and concrete applications of artificial intelligence in libraries have been explored and published for decades. Over time, faster processing allows use at scale. Second, how library and metadata work contributes to machine …

Instance-Specific Algorithm Configuration Via Unsupervised Deep Graph Clustering, Wen Song, Yi Liu, Zhiguang Cao, Yaoxin Wu, Qiqiang Li

Research Collection School Of Computing and Information Systems

Instance-specific Algorithm Configuration (AC) methods are effective in automatically generating high-quality algorithm parameters for heterogeneous NP-hard problems from multiple sources. However, existing works rely on manually designed features to describe training instances, which are simple numerical attributes and cannot fully capture structural differences. Targeting at Mixed-Integer Programming (MIP) solvers, this paper proposes a novel instances-specific AC method based on end-to-end deep graph clustering. By representing an MIP instance as a bipartite graph, a random walk algorithm is designed to extract raw features with both numerical and structural information from the instance graph. Then an auto-encoder is designed to learn dense …

Multi-Representation Variational Autoencoder Via Iterative Latent Attention And Implicit Differentiation, Nhu Thuat Tran, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Variational Autoencoder (VAE) offers a non-linear probabilistic modeling of user's preferences. While it has achieved remarkable performance at collaborative filtering, it typically samples a single vector for representing user's preferences, which may be insufficient to capture the user's diverse interests. Existing solutions extend VAE to model multiple interests of users by resorting a variant of self-attentive method, i.e., employing prototypes to group items into clusters, each capturing one topic of user's interests. Despite showing improvements, the current design could be more effective since prototypes are randomly initialized and shared across users, resulting in uninformative and non-personalized clusters.To fill the gap, …

Deep Reinforcement Learning With Explicit Context Representation, Francisco Munguia-Galeano, Ah-Hwee Tan, Ze Ji

Research Collection School Of Computing and Information Systems

Though reinforcement learning (RL) has shown an outstanding capability for solving complex computational problems, most RL algorithms lack an explicit method that would allow learning from contextual information. On the other hand, humans often use context to identify patterns and relations among elements in the environment, along with how to avoid making wrong actions. However, what may seem like an obviously wrong decision from a human perspective could take hundreds of steps for an RL agent to learn to avoid. This article proposes a framework for discrete environments called Iota explicit context representation (IECR). The framework involves representing each state …

Carbon-Aware Mine Planning With A Novel Multi-Objective Framework, Nurul Asyikeen Binte Azhar, Aldy Gunawan, Shih-Fen Cheng, Erwin Leonardi

Research Collection School Of Computing and Information Systems

The logistical complication of long-term mine planning involves deciding the sequential extraction of materials from the mine pit and their subsequent processing steps based on geological, geometrical, and resource constraints. The net present value (NPV) of profit over the mine's lifespan usually forms the sole objective for this problem, which is considered as the NP-hard precedence-constrained production scheduling problem (PCPSP) as well. However, increased pressure for more sustainable and carbon-aware industries also calls for environmental indicators to be considered. In this paper, we enhance the generic PCPSP formulation into a multi-objective optimization (MOO) problem whereby carbon cost forms an additional …

Uncertainty-Adjusted Inductive Matrix Completion With Graph Neural Networks, Petr Kasalicky, Antoine Ledent, Rodrigo Alves

Research Collection School Of Computing and Information Systems

We propose a robust recommender systems model which performs matrix completion and a ratings-wise uncertainty estimation jointly. Whilst the prediction module is purely based on an implicit low-rank assumption imposed via nuclear norm regularization, our loss function is augmented by an uncertainty estimation module which learns an anomaly score for each individual rating via a Graph Neural Network: data points deemed more anomalous by the GNN are downregulated in the loss function used to train the low-rank module. The whole model is trained in an end-to-end fashion, allowing the anomaly detection module to tap on the supervised information available in …

Quantifying Taxi Drivers' Behaviors With Behavioral Game Theory, Mengyu Ji, Yuhong Xu, Shih-Fen Cheng

Research Collection School Of Computing and Information Systems

With their flexibility and convenience, taxis play a vital role in urban transportation systems. Understanding how human drivers make decisions in a context of uncertainty and competition is crucial for taxi fleets that depend on drivers to provide their services. As part of this paper, we propose modeling taxi drivers’ behaviors based on behavioral game theory. Based on real-world data, we demonstrate that the behavioral game theory model we select is superior to state-of-the-art baselines. These results provide a solid foundation for improving taxi fleet efficiency in the future.

Learning To Send Reinforcements: Coordinating Multi-Agent Dynamic Police Patrol Dispatching And Rescheduling Via Reinforcement Learning, Waldy Joe, Hoong Chuin Lau

Research Collection School Of Computing and Information Systems

We address the problem of coordinating multiple agents in a dynamic police patrol scheduling via a Reinforcement Learning (RL) approach. Our approach utilizes Multi-Agent Value Function Approximation (MAVFA) with a rescheduling heuristic to learn dispatching and rescheduling policies jointly. Often, police operations are divided into multiple sectors for more effective and efficient operations. In a dynamic setting, incidents occur throughout the day across different sectors, disrupting initially-planned patrol schedules. To maximize policing effectiveness, police agents from different sectors cooperate by sending reinforcements to support one another in their incident response and even routine patrol. This poses an interesting research challenge …

Document-Level Relation Extraction Via Separate Relation Representation And Logical Reasoning, Heyan Huang, Changsen Yuan, Qian Liu, Yixin Cao

Research Collection School Of Computing and Information Systems

Document-level relation extraction (RE) extends the identification of entity/mentions’ relation from the single sentence to the long document. It is more realistic and poses new challenges to relation representation and reasoning skills. In this article, we propose a novel model, SRLR, using Separate Relation Representation and Logical Reasoning considering the indirect relation representation and complex reasoning of evidence sentence problems. Specifically, we first expand the judgment of relational facts from the entity-level to the mention-level, highlighting fine-grained information to capture the relation representation for the entity pair. Second, we propose a logical reasoning module to identify evidence sentences and conduct …

Techsumbot: A Stack Overflow Answer Summarization Tool For Technical Query, Chengran Yang, Bowen Xu, Jiakun Liu, David Lo

Research Collection School Of Computing and Information Systems

Stack Overflow is a popular platform for developers to seek solutions to programming-related problems. However, prior studies identified that developers may suffer from the redundant, useless, and incomplete information retrieved by the Stack Overflow search engine. To help developers better utilize the Stack Overflow knowledge, researchers proposed tools to summarize answers to a Stack Overflow question. However, existing tools use hand-craft features to assess the usefulness of each answer sentence and fail to remove semantically redundant information in the result. Besides, existing tools only focus on a certain programming language and cannot retrieve up-to-date new posted knowledge from Stack Overflow. …

Investigating The Use Of Recurrent Neural Networks In Modeling Guitar Distortion Effects, Caleb Koch, Scott Hawley, Andrew Fyfe

Belmont University Research Symposium (BURS)

Guitar players have been modifying their guitar tone with audio effects ever since the mid-20th century. Traditionally, these effects have been achieved by passing a guitar signal through a series of electronic circuits which modify the signal to produce the desired audio effect. With advances in computer technology, audio “plugins” have been created to produce audio effects digitally through programming algorithms. More recently, machine learning researchers have been exploring the use of neural networks to replicate and produce audio effects initially created by analog and digital effects units. Recurrent Neural Networks have proven to be exceptional at modeling audio effects …

Chatgpt As Metamorphosis Designer For The Future Of Artificial Intelligence (Ai): A Conceptual Investigation, Amarjit Kumar Singh (Library Assistant), Dr. Pankaj Mathur (Deputy Librarian)

Library Philosophy and Practice (e-journal)

Abstract

Purpose: The purpose of this research paper is to explore ChatGPT’s potential as an innovative designer tool for the future development of artificial intelligence. Specifically, this conceptual investigation aims to analyze ChatGPT’s capabilities as a tool for designing and developing near about human intelligent systems for futuristic used and developed in the field of Artificial Intelligence (AI). Also with the helps of this paper, researchers are analyzed the strengths and weaknesses of ChatGPT as a tool, and identify possible areas for improvement in its development and implementation. This investigation focused on the various features and functions of ChatGPT that …

Regulating Machine Learning: The Challenge Of Heterogeneity, Cary Coglianese

All Faculty Scholarship

Machine learning, or artificial intelligence, refers to a vast array of different algorithms that are being put to highly varied uses, including in transportation, medicine, social media, marketing, and many other settings. Not only do machine-learning algorithms vary widely across their types and uses, but they are evolving constantly. Even the same algorithm can perform quite differently over time as it is fed new data. Due to the staggering heterogeneity of these algorithms, multiple regulatory agencies will be needed to regulate the use of machine learning, each within their own discrete area of specialization. Even these specialized expert agencies, though, …

Layout Generation As Intermediate Action Sequence Prediction, Huiting Yang, Danqing Huang, Chin-Yew Lin, Shengfeng He

Research Collection School Of Computing and Information Systems

Layout generation plays a crucial role in graphic design intelligence. One important characteristic of the graphic layouts is that they usually follow certain design principles. For example, the principle of repetition emphasizes the reuse of similar visual elements throughout the design. To generate a layout, previous works mainly attempt at predicting the absolute value of bounding box for each element, where such target representation has hidden the information of higher-order design operations like repetition (e.g. copy the size of the previously generated element). In this paper, we introduce a novel action schema to encode these operations for better modeling the …

Generalizing Math Word Problem Solvers Via Solution Diversification, Zhenwen Liang, Jipeng Zhang, Lei Wang, Yan Wang, Jie Shao, Xiangliang Zhang

Research Collection School Of Computing and Information Systems

Current math word problem (MWP) solvers are usually Seq2Seq models trained by the (one-problem; one-solution) pairs, each of which is made of a problem description and a solution showing reasoning flow to get the correct answer. However, one MWP problem naturally has multiple solution equations. The training of an MWP solver with (one-problem; one-solution) pairs excludes other correct solutions, and thus limits the generalizability of the MWP solver. One feasible solution to this limitation is to augment multiple solutions to a given problem. However, it is difficult to collect diverse and accurate augment solutions through human efforts. In this paper, …

Scalable And Globally Optimal Generalized L1 K-Center Clustering Via Constraint Generation In Mixed Integer Linear Programming, Aravinth Chembu, Scott Sanner, Hassan Khurran, Akshat Kumar

Research Collection School Of Computing and Information Systems

The k-center clustering algorithm, introduced over 35 years ago, is known to be robust to class imbalance prevalent in many clustering problems and has various applications such as data summarization, document clustering, and facility location determination. Unfortunately, existing k-center algorithms provide highly suboptimal solutions that can limit their practical application, reproducibility, and clustering quality. In this paper, we provide a novel scalable and globally optimal solution to a popular variant of the k-center problem known as generalized L_1 k-center clustering that uses L_1 distance and allows the selection of arbitrary vectors as cluster centers. We show that this clustering objective …

Architectural Design Of A Blockchain-Enabled, Federated Learning Platform For Algorithmic Fairness In Predictive Health Care: Design Science Study, Xueping Liang, Juan Zhao, Yan Chen, Eranga Bandara, Sachin Shetty

VMASC Publications

Background: Developing effective and generalizable predictive models is critical for disease prediction and clinical decision-making, often requiring diverse samples to mitigate population bias and address algorithmic fairness. However, a major challenge is to retrieve learning models across multiple institutions without bringing in local biases and inequity, while preserving individual patients' privacy at each site.

Objective: This study aims to understand the issues of bias and fairness in the machine learning process used in the predictive health care domain. We proposed a software architecture that integrates federated learning and blockchain to improve fairness, while maintaining acceptable prediction accuracy and minimizing overhead …

Patch-Wise Training With Convolutional Neural Networks To Synthetically Upscale Cfd Simulations, John P. Romano, Alec C. Brodeur, Oktay Baysal

Mechanical & Aerospace Engineering Faculty Publications

This paper expands the authors’ prior work[1], which focuses on developing a convolutional neural network (CNN) model capable of mapping time-averaged, unsteady Reynold’s-averaged Navier-Stokes (URANS) simulations to higher resolution results informed by time-averaged detached eddy simulations (DES). The authors present improvements over the prior CNN autoencoder model that result from hyperparameter optimization, increased data set augmentation through the adoption of a patch-wise training approach, and the predictions of primitive variables rather than vorticity magnitude. The training of the CNN model developed in this study uses the same URANS and DES simulations of a transonic flow around several NACA 4-digit airfoils …

A Structure-Aware Generative Adversarial Network For Bilingual Lexicon Induction, Bocheng Han, Qian Tao, Lusi Li, Zhihao Xiong

Computer Science Faculty Publications

Bilingual lexicon induction (BLI) is the task of inducing word translations with a learned mapping function that aligns monolingual word embedding spaces in two different languages. However, most previous methods treat word embeddings as isolated entities and fail to jointly consider both the intra-space and inter-space topological relations between words. This limitation makes it challenging to align words from embedding spaces with distinct topological structures, especially when the assumption of isomorphism may not hold. To this end, we propose a novel approach called the Structure-Aware Generative Adversarial Network (SA-GAN) model to explicitly capture multiple topological structure information to achieve accurate …

Dynamic Function Learning Through Control Of Ensemble Systems, Wei Zhang, Vignesh Narayanan, Jr-Shin Li

Publications

Learning tasks involving function approximation are preva- lent in numerous domains of science and engineering. The underlying idea is to design a learning algorithm that gener- ates a sequence of functions converging to the desired target function with arbitrary accuracy by using the available data samples. In this paper, we present a novel interpretation of iterative function learning through the lens of ensemble dy- namical systems, with an emphasis on establishing the equiv- alence between convergence of function learning algorithms and asymptotic behavior of ensemble systems. In particular, given a set of observation data in a function learning task, we …

Cooperative Deep Q -Learning Framework For Environments Providing Image Feedback, Krishnan Raghavan, Vignesh Narayanan, Sarangapani Jagannathan

Publications

In this article, we address two key challenges in deep reinforcement learning (DRL) setting, sample inefficiency, and slow learning, with a dual-neural network (NN)-driven learning approach. In the proposed approach, we use two deep NNs with independent initialization to robustly approximate the action-value function in the presence of image inputs. In particular, we develop a temporal difference (TD) error-driven learning (EDL) approach, where we introduce a set of linear transformations of the TD error to directly update the parameters of each layer in the deep NN. We demonstrate theoretically that the cost minimized by the EDL regime is an approximation …