Artificial Intelligence and Robotics Commons^™

Documenting Ethical Considerations In Open Source Ai Models, Haoyu Gao, Mansooreh Zahedi, Christoph Treude, Sarita Rosenstock, Marc Cheong

Research Collection School Of Computing and Information Systems

Background: The development of AI-enabled software heavily depends on AI model documentation, such as model cards, due to different domain expertise between software engineers and model developers. From an ethical standpoint, AI model documentation conveys critical information on ethical considerations along with mitigation strategies for downstream developers to ensure the delivery of ethically compliant software. However, knowledge on such documentation practice remains scarce. Aims: The objective of our study is to investigate how developers document ethical aspects of open source AI models in practice, aiming at providing recommendations for future documentation endeavours. Method: We selected three sources of documentation on …

Self-Supervised Learning For Time Series Analysis : Taxonomy, Progress, And Prospects, Zhang Kexin, Qingsong Wen, Chaoli Zhang, Rongyao Cai, Ming Jin, Yong Liu, James Y. Zhang, Guansong Pang, Guansong Pang, Pan Shirui

Research Collection School Of Computing and Information Systems

Self-supervised learning (SSL) has recently achieved impressive performance on various time series tasks. The most prominent advantage of SSL is that it reduces the dependence on labeled data. Based on the pre-training and fine-tuning strategy, even a small amount of labeled data can achieve high performance. Compared with many published self-supervised surveys on computer vision and natural language processing, a comprehensive survey for time series SSL is still missing. To fill this gap, we review current state-of-the-art SSL methods for time series data in this article. To this end, we first comprehensively review existing surveys related to SSL and time …

Latent Representation Learning For Geospatial Entities, Ween Jiann Lee, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Representation learning has been instrumental in the success of machine learning, offering compact and performant data representations for diverse downstream tasks. In the spatial domain, it has been pivotal in extracting latent patterns from various data types, including points, polylines, polygons, and networked structures. However, existing approaches often fall short of explicitly capturing both semantic and spatial information, relying on proxies and synthetic features. This article presents GeoNN, a novel graph neural network-based model designed to learn spatially-aware embeddings for geospatial entities. GeoNN leverages edge features generated from geodesic functions, dynamically selecting relevant features based on relative locations. It introduces …

Promise And Peril Of Collaborative Code Generation Models : Balancing Effectiveness And Memorization, Zhi Chen, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

In the rapidly evolving field of machine learning, training models with datasets from various locations and organizations presents significant challenges due to privacy and legal concerns. The exploration of effective collaborative training settings, which are capable of leveraging valuable knowledge from distributed and isolated datasets, is increasingly crucial. This study investigates key factors that impact the effectiveness of collaborative training methods in code next-token prediction, as well as the correctness and utility of the generated code, showing the promise of such methods. Additionally, we evaluate the memorization of different participant training data across various collaborative training settings, including centralized, federated, …

Constrained Assortment Optimization Under The Cross-Nested Logit Model, Cuong Le, Tien Mai

Research Collection School Of Computing and Information Systems

We study the assortment optimization problem under general linear constraints, where the customer choice behavior is captured by the cross-nested logit model. In this problem, there is a set of products organized into multiple subsets (or nests), where each product can belong to more than one nest. The aim is to find an assortment to offer to customers so that the expected revenue is maximized. We show that, under the cross-nested logit model, the unconstrained assortment problem is NP-hard even when there are only two nests, and the problem is generally NP-hard to approximate to any constant factors. To tackle …

Themis: Automatic And Efficient Deep Learning System Testing With Strong Fault Detection Capability, Dong Huang, Tsz On Li, Xiaofei Xie, Heming Cui

Research Collection School Of Computing and Information Systems

Deep Learning Systems (DLSs) have been widely applied in safety-critical tasks such as autopilot. However, when a perturbed input is fed into a DLS for inference, the DLS often has incorrect outputs (i.e., faults). DLS testing techniques (e.g., DeepXplore) detect such faults by generating perturbed inputs to explore data flows that induce faults. Since a DLS often has infinitely many data flows, existing techniques require developers to manually specify a set of activation values in a DLS’s neurons for exploring fault-inducing data flows. Unfortunately, recent studies show that such manual effort is tedious and can detect only a tiny proportion …

Joint Weakly Supervised Image Emotion Analysis Based On Interclass Discrimination And Intraclass Correlation, Xinyue Zhang, Zhaoxia Wang, Guitao Cao, Seng-Beng Ho

Research Collection School Of Computing and Information Systems

Regional information-based image emotion analysis has recently garnered significant attention. However, existing methods often focus on identifying region proposals through layered steps or merely rely on visual saliency. These approaches may lead to an underestimation of emotional categories and a lack of comprehensive interclass discrimination perception and emotional intraclass contextual mining. To address these limitations, we propose a novel approach named InterIntraIEA, which combines interclass discrimination and intraclass correlation joint learning capabilities for image emotion analysis. The proposed method not only employs category-specific dictionary learning for class adaptation, but also models intraclass contextual relationships and perceives correlations at the channel …

Navigating Governance Paradigms: A Cross-Regional Comparative Study Of Generative Ai Governance Processes & Principle, Jose Luna, Ivan Tan, Xiaofei Xie, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

As Generative Artificial Intelligence (GenAI) technologies evolve at an unprecedented rate, global governance approaches struggle to keep pace with the technology, highlighting a critical issue in the governance adaptation of significant challenges. Depicting the nuances of nascent and diverse governance approaches based on risks, rules, outcomes, principles, or a mix across different regions around the globe is fundamental to discern discrepancies and convergences and to shed light on specific limitations that need to be addressed, thereby facilitating the safe and trustworthy adoption of GenAI. In response to the need and the evolving nature of GenAI, this paper seeks to provide …

Exploring Conversations Between A Practitioner And A Person With Dementia, Kotaro Hara, Rosiana Natalie, Wei Soon Cheong, Jingjing Gu, Qianli Xu

Research Collection School Of Computing and Information Systems

In social service centers, practitioners engage in conversations with clients with dementia to facilitate their daily activities and provide support when they are distressed. However, the nature of the care demands the practitioner’s active engagement, which becomes difficult to deliver as the number of people who need care expands. Researchers have been investigating the efficacy of developing agents that assume conversational tasks to alleviate this work. To contribute to the future design of agents for caregiving, we collected and analyzed ten conversations between clients with mild dementia and practitioners who provide care. Our analyses of turn-taking dynamics and dialogue acts …

Efficient Cascaded Multiscale Adaptive Network For Image Restoration, Yichen Zhou, Pan Zhou, Teck Khim Ng

Research Collection School Of Computing and Information Systems

Image restoration, encompassing tasks such as deblurring, denoising, and super-resolution, remains a pivotal area in computer vision. However, efficiently addressing the spatially varying artifacts of various low-quality images with local adaptiveness and handling their degradations at different scales poses significant challenges. To efficiently tackle these issues, we propose the novel Efficient Cascaded Multiscale Adaptive (ECMA) Network. ECMA employs Local Adaptive Module, LAM, which dynamically adjusts convolution kernels across local image regions to efficiently handle varying artifacts. Thus, LAM addresses the local adaptiveness challenge more efficiently than costlier mechanisms like self-attention, due to its less computationally intensive convolutions. To construct a …

Can Federated Learning Solve Ai’S Data Privacy Problem?: A Legal Analysis, Warren B. Chik, Florian Gamper

Research Collection Yong Pung How School Of Law

Federated learning (FL) is a method of training AI systems on different datasets without sharing data. The promise of FL is to enable AI systems to be trained on data, including personal data, while preserving data privacy and confidentiality, and thus, inter alia, facilitate compliance with data protection legislation. FL has generated a considerable interest amongst the computer science community, yet there is a dearth of legal analysis of FL. This is a problem because the question of whether FL facilitates compliance with data protection legislation is a legal question. This article will fill this lacuna by providing a comprehensive …

Trust And Robotics: A Multi-Staged Decision-Making Approach To Robots In Community, Wenxi Zhang, Willow Wong, Mark Findlay

Research Collection Yong Pung How School Of Law

With the desired outcome of social good within the wider robotics ecosystem, trust is identified as the central adhesive of the human–robot interaction (HRI) interface. However, building trust between humans and robots involves more than improving the machine’s technical reliability or trustworthiness in function. This paper presents a holistic, community-based approach to trust-building, where trust is understood as a multifaceted and multi-staged looped relation that depends heavily on context and human perceptions. Building on past literature that identifies dispositional and learned stages of trust, our proposed decision to trust model considers more extensively the human and situational factors influencing how …

The Promise And Perils Of Smart (City) Bots As Educational Tools, Thomas Menkhoff, Siew Ning Kan, Shaohui Foong

Research Collection Lee Kong Chian School Of Business

Written from an educator’s perspective, the paper aims to examine the increasing use of chatbot technology capable of answering user queries in various fields such as commerce, public service delivery, urban areas, higher education etc. by simulating human conversation through text messages or voice commands. We analyse why smart bots have become popular technologies in the context smart cities; shed light on different types of bots (chatbots, large language models such as ChatGPT, flying bots); and explain how we use bots as teaching tool in an introductory AI for business course. A key argument putting forward is that smart bots …

Leveraging Imitation Learning In Agricultural Robotics: A Comprehensive Survey And Comparative Analysis, Siavash Mahmoudi, Amirreza Davar, Pouya Sohrabipour, Ramesh Bahadur Bist, Yang Tao, Dongyi Wang

Biological and Agricultural Engineering Faculty Publications and Presentations

Imitation learning (IL), a burgeoning frontier in machine learning, holds immense promise across diverse domains. In recent years, its integration into robotics has sparked significant interest, offering substantial advancements in autonomous control processes. This paper presents an exhaustive insight focusing on the implementation of imitation learning techniques in agricultural robotics. The survey rigorously examines varied research endeavors utilizing imitation learning to address pivotal agricultural challenges. Methodologically, this survey comprehensively investigates multifaceted aspects of imitation learning applications in agricultural robotics. The survey encompasses the identification of agricultural tasks that can potentially be addressed through imitation learning, detailed analysis of specific models …

Robust And Generalizable Representations In The Hippocampus, Hung-Tu Chen

Dartmouth College Ph.D Dissertations

An intelligent system must balance generalizing across similar experiences with maintaining the distinctiveness of each experience. This thesis explores how the hippocampus manages this balance through its neural representations to support adaptive behavior. In Chapter 1, I provide an overview of key hippocampal phenomena that contribute to this process, including remapping, splitter signal, and replay. In Chapter 2, I challenge the concept of random remapping by showing that it is possible to predict, better than chance, how a given experience will be encoded in the hippocampus across different subjects. This suggests that encoding of related experiences, which was previously thought …

Privacy Risks And Regulatory Challenges In Smart Grids And Renewable Energy Systems, Mikołaj Rajca

internetowy Kwartalnik Antymonopolowy i Regulacyjny (internet Quarterly on Antitrust and Regulation)

Smart grid technologies are central to the global shift towards a more efficient and sustainable energy infrastructure, integrating advanced digital systems with traditional power networks. While these technologies offer significant benefits, including enhanced energy management and the seamless integration of renewable energy sources, they also introduce complex privacy challenges. The extensive data collection and real-time communication capabilities inherent in smart grids raise concerns over consumer privacy, data breaches, and cybersecurity threats. This paper critically examines these privacy risks within the context of evolving regulatory frameworks such as the GDPR, NIS2 Directive, and the forthcoming EU AI Act. The discussion emphasizes …

A Framework For Human Evaluation Of Large Language Models In Healthcare Derived From Literature Review, Thomas Yu Chow Tam, Sonish Sivarajkumar, Sumit Kapoor, Alisa V Stolyar, Katelyn Polanska, Karleigh R Mccarthy, Hunter Osterhoudt, Xizhi Wu, Shyam Visweswaran, Sunyang Fu, Piyush Mathur, Giovanni E Cacciamani, Cong Sun, Yifan Peng, Yanshan Wang

Faculty, Staff and Student Publications

With generative artificial intelligence (GenAI), particularly large language models (LLMs), continuing to make inroads in healthcare, assessing LLMs with human evaluations is essential to assuring safety and effectiveness. This study reviews existing literature on human evaluation methodologies for LLMs in healthcare across various medical specialties and addresses factors such as evaluation dimensions, sample types and sizes, selection, and recruitment of evaluators, frameworks and metrics, evaluation process, and statistical analysis type. Our literature review of 142 studies shows gaps in reliability, generalizability, and applicability of current human evaluation practices. To overcome such significant obstacles to healthcare LLM developments and deployments, we …

Architecting Standards: Leveraging Ia To Drive Emerging Technology Standards​, Noreen Y. Whysel

Publications and Research

In this talk for the annual meeting of the W3C Information Architecture Community Group, I discuss how standards bodies can learn from the Information Architecture field to improve accessibility and usability. By adopting user-centered design principles and engaging in iterative feedback loops, standards organizations can create more relevant and practical standards that align with real-world applications. Emphasizing collaboration with industry experts and end-users will help ensure that standards evolve along with technological advancements.

In the meantime, key standards are examined, such as ISO 9241, which provides guidelines for usability and user-centered design, and the W3C's Web Content Accessibility Guidelines (WCAG), …

The Limits Of Generative Ai In Administrative Law Research, Susan Azyndar

Journal Articles

The author recounts an administrative law classroom experience using generative AI. She considers the complexities of administrative law, AI training, professional responsibility, and traditional resources.

Revolutionizing Public Safety And Criminal Justice Through Ai, Alan Saquella

Publications

Artificial Intelligence (AI) is rapidly transforming public safety, criminal justice and security by fundamentally changing how crimes are committed, investigated and prevented. As AI tools become increasingly sophisticated, law enforcement and corporate security professionals are utilizing these advancements to enhance their capabilities. However, integrating AI into these sectors also brings significant challenges, including ethical concerns, recruitment difficulties, and the surge in crime rates. This article examines the transformative impact of AI, the ongoing efforts to unify AI applications across public safety and security sectors, and expert advice on overcoming the associated challenges.