Digital Commons Network^™

Robust Asynchronous Federated Learning With Time-Weighted And Stale Model Aggregation, Yinbin Miao, Ziteng Liu, Xinghua Li, Meng Li, Hongwei Li, Kim-Kwang Raymond Choo, Robert H. Deng

Research Collection School Of Computing and Information Systems

Federated Learning (FL) ensures collaborative learning among multiple clients while maintaining data locally. However, the traditional synchronous FL solutions have lower accuracy and require more communication time in scenarios where most devices drop out during learning. Therefore, we propose an Asynchronous Federated Learning (AsyFL) scheme using time-weighted and stale model aggregation, which effectively solves the problem of poor model performance due to the heterogeneity of devices. Then, we integrate Symmetric Homomorphic Encryption (SHE) into AsyFL to propose Asynchronous Privacy-Preserving Federated Learning (Asy-PPFL), which protects the privacy of clients and achieves lightweight computing. Privacy analysis shows that Asy-PPFL is indistinguishable under …

Empathyear : An Open-Source Avatar Multimodal Empathetic Chatbot, Hao Fei, Han Zhang, Bin Wang, Lizi Liao, Qian Liu, Erik Cambria

Research Collection School Of Computing and Information Systems

This paper introduces EmpathyEar, a pioneering open-source, avatar-based multimodal empathetic chatbot, to fill the gap in traditional text-only empathetic response generation (ERG) systems. Leveraging the advancements of a large language model, combined with multimodal encoders and generators, EmpathyEar supports user inputs in any combination of text, sound, and vision, and produces multimodal empathetic responses, offering users, not just textual responses but also digital avatars with talking faces and synchronized speeches. A series of emotion-aware instruction-tuning is performed for comprehensive emotional understanding and generation capabilities. In this way, EmpathyEar provides users with responses that achieve a deeper emotional resonance, closely emulating …

Is This Home Truly? Reflections On Singapore As It Turns 59, Lily Kong

Research Collection School of Social Sciences

In a commentary, SMU President Professor Lily Kong shared her reflections on Singapore as it turns 59. She discussed Singapore's nation-building process and offered her views on how the country can strengthen its social compact to make Singapore a home for Singaporeans.

My Manager Endorsed My Coworkers’ Voice: Understanding Observers’ Positive And Negative Reactions To Managerial Endorsement Of Coworker Voice., Emily Poulton, Szu-Han Joanna Lin, Shereen Fatimah, Cony Ho, Lance Ferris, Russell Johnson

Research Collection Lee Kong Chian School Of Business

Research on managerial voice endorsement has primarily focused on the processes and conditions through which voicers receive their managers’ endorsement. We shift this focus away from the voicers, focusing instead on the dual reactions that endorsement generates for observing employees. Drawing from an approach-avoidance framework, we propose that managerial endorsement of coworker voice could be perceived as a positive and negative stimulus for observers, prompting them to approach opportunities and avoid threats, respectively. Results from a pre-registered experiment and a multi-wave, multi-source field study revealed that managerial endorsement of coworker voice was positively related to observers’ voice instrumentality, thus prompting …

Digitally Connected, Evolutionarily Wired: An Evolutionary Mismatch Perspective On Digital Work, Mark Van Vugt, Stephen M. Colarelli, Norman P. Li

Research Collection School of Social Sciences

This paper makes the case for an evolutionary mismatch between digital work and the way human ancestors engaged in work. Psychological adaptations for producing things that early humans needed to survive and thrive, such as cognitive mechanisms for obtaining and processing food, toolmaking, and learning valuable working skills, evolved in the context of small networks of hunter–gatherers. These adaptations are central to understanding the significance of work in human evolution. Evolutionary mismatches operate when novel environments cue ancestral adaptations in ways that no longer provide adaptive benefits. We argue that digital work, although efficient and productive, is misaligned with some …

Post-Productivism And Rural Revitalization In China: Drivers And Outcomes, Meiling Wu, Qian Forrest Zhang, John A. Donaldson

Research Collection School of Social Sciences

We propose that post-productivism offers a useful analytical framework for understanding the multi-scalar and diverse changes that are taking place in China’s rural revitalization. As a theoretical framework that emerged from the study of rural changes in the Global North, the applicability of post-productivism in the Global South has been contested. This paper offers the first comprehensive analysis of the emergence of post-productivism in rural China and uses post-productivism as a framework to conceptualize a wide range of changes in China’s rural revitalization. We conceptually clarify the driving forces that give rise to post-productivism and the outcomes these drivers produce. …

Exploring A Multimodal Fusion-Based Deep Learning Network For Detecting Facial Palsy, Heng Yim Nicole Oo, Min Hun Lee, J. H. Lim

Research Collection School Of Computing and Information Systems

Algorithmic detection of facial palsy offers the potential to improve current practices, which usually involve labor-intensive and subjective assessment by clinicians. In this paper, we present a multimodal fusion-based deep learning model that utilizes unstructured data (i.e. an image frame with facial line segments) and structured data (i.e. features of facial expressions) to detect facial palsy. We then contribute to a study to analyze the effect of different data modalities and the benefits of a multimodal fusion-based approach using videos of 21 facial palsy patients. Our experimental results show that among various data modalities (i.e. unstructured data - RGB images …

Geographic Links And Predictable Returns, Zuben Jin, Frank Weikai Li

Research Collection Lee Kong Chian School Of Business

Using establishment-level data of U.S. public firms, we construct a novel measure of geographic linkage between firms. We show that the returns of geography-linked firms have strong predictive power for focal firm returns and fundamentals. This effect is distinct from other cross-firm return predictability and is not easily attributable to risk-based explanations. It is more pronounced for focal firms that receive lower investor attention, are more costly to arbitrage, and during high sentiment periods. The cross-firm information spillovers and return predictability are also stronger for geographic peers with economic linkages and with positive information. Our results are broadly consistent with …

A Theory Of Host Country Sentiments: An Illustration In Cross-Border Acquisitions, Daphne W. Yiu, William P. Wan, Xiaocong Tian, Kelly Xing Chen

Research Collection Lee Kong Chian School Of Business

Capitalizing on massive sentiment diffusion and mobilization aided by mass media and social media nowadays, we introduce a general theory of host country sentiments to illuminate the granular, context-situated, and time dynamic power of social sentiments. We posit that host country is a value-laden context wherein a foreign entity’s strategic action stimulates host country stakeholders’ social sentiments and engagement in systems of discourses (“sentiment arousal”), mobilizes social sentiments toward the foreign entity (“sentiment competence”), and legitimizes social sentiments and the strategic action of the foreign entity (“sentiment resonance”). To test our theory, we employ a context of inward acquisitions by …

Predicting Personality Or Prejudice? Facial Inference In The Age Of Artificial Intelligence, Shilpa Madan, Gayoung Park

Research Collection Lee Kong Chian School Of Business

Facial inference, a cornerstone of person perception, has traditionally been studied through human judgments about personality traits and abilities based on people's faces. Recent advances in artificial intelligence (AI) have introduced new dimensions to this field, employing machine learning algorithms to reveal people's character, capabilities, and social outcomes based just on their faces. This review examines recent research on human and AI-based facial inference across psychology, business, computer science, legal, and policy studies to highlight the need for scientific consensus on whether or not people's faces can reveal their inner traits, and urges researchers to address the critical concerns …

Research On Digital Platform Capabilities And Entrepreneurship Empowerment Performance Of Science And Technology Industrial Parks, Jieyu Zheng

Dissertations and Theses Collection (Open Access)

In the era of the digital economy, industries are shifting toward platform-based and ecological trends, with industrial parks similarly adopting digital and platform-based patterns. The systematic planning, functional layout, and service efficiency of science and technology industrial parks (sci-tech industrial parks for short) are designed to address emerging industries’ spatial and organizational needs. This approach aims to create a system where digital transformation is employed as the mechanism and digital platforms empower entrepreneurial enterprises within these parks, fostering high-quality park development. However, despite efforts to build digital platforms for transforming into digital parks, many industrial parks have found their platform …

Essays On Aging, Health, And Labor Market, Lanjie Wang

Dissertations and Theses Collection (Open Access)

This dissertation consists of four chapters on aging, health, and labor economics.

In the first chapter, we investigate the short-term impacts of unemployment on older workers in Singapore, a setting without public unemployment insurance. Using monthly panel data, we analyze dynamic impacts on income, spending, health, and subjective well-being over two years post-unemployment. We find substantial initial earnings losses with incomplete recovery, with income remaining 46.5% below pre-unemployment levels after 24 months. Despite this persistent income gap, consumption responses are modest and comparable to countries with strong social safety nets, suggesting robust self-insurance mechanisms. We observe increased retirement among married …

Deep Representation Learning For Time Series Forecasting, Gerald Woo

Dissertations and Theses Collection (Open Access)

Time series forecasting has critical applications across business and scien- tific domains, such as demand forecasting, capacity planning and management, and anomaly detection. Being able to predict the future yields immense value, allowing us to make downstream decisions with more confidence. Deep learning for time series forecasting is a burgeoning area of research, moving away from simple linear models found in classical time series analysis literature, towards more expressive, data hungry neural network architectures.

In this thesis, we develop methods leveraging deep representation learning for time series forecasting, from exploring neural network architecture designs which encode inductive biases specific to …

Digital Payments And Consumption: Evidence From The 2016 Demonetization In India, Sumit Agarwal, Pulak Ghosh, Jing Li, Tianyue Ruan

Research Collection School Of Economics

We study how consumer spending responds to digital payments, using the differential switch to digital payments across consumers induced by the sudden 2016 Indian Demonetization for identification. Usage of digital payments rose by 3.38 percentage points and monthly spending increased by 3% for an additional 10 percentage points in prior cash dependence. Spending remained elevated even when cash availability recovered. Robustness analyses show that the spending response is not driven by income shocks, credit supply, price changes, or consumers' moving to the formal market. We provide evidence that digital payments increase consumer spending due to subdued salience.

Value Capture And Affordable Housing: Insights From Singapore, Sock Yong Phang

Research Collection School Of Economics

To explain the role of value capture in Singapore’s progress, I will first provide some background information on Singapore and its housing sector. I will then take you on a journey back to the 1960s to show how we established our present institutional arrangements for land and housing. I will next describe how housing policies have evolved over the past six decades in response to changing contexts. Finally, I will conclude with the fiscal implications of successful value capture. Where relevant, I will indicate how Singapore’s policies align with George’s views.

A Robust Optimization Approach To Mechanism Design, Jiangtao Li, Kexin Wang

Research Collection School Of Economics

We study the design of mechanisms when the mechanism designer faces local uncertainty about agents’ beliefs. Specifically, we consider a designer who does not know the exact beliefs of the agents but is confident that her estimate is within ϵ of the beliefs held by the agents (where ϵ reflects the degree of local uncertainty). Adopting the robust optimization approach, we design mechanisms that incentivize agents to truthfully report their payoff-relevant information regardless of their actual beliefs. For any fixed ϵ, we identify necessary and sufficient conditions under which requiring this sense of robustness is without loss of revenue for …

Gradient Wild Bootstrap For Instrumental Variable Quantile Regressions With Weak And Few Clusters, Wenjie Wang, Yichong Zhang

Research Collection School Of Economics

We study the gradient wild bootstrap-based inference for instrumental variable quantile regressions in the framework of a small number of large clusters in which the number of clusters is viewed as fixed, and the number of observations for each cluster diverges to infinity. For the Wald inference, we show that our wild bootstrap Wald test, with or without studentization using the cluster-robust covariance estimator (CRVE), controls size asymptotically up to a small error as long as the parameter of endogenous variable is strongly identified in at least one of the clusters. We further show that the wild bootstrap Wald test …

Topic Modeling On Document Networks With Dirichlet Optimal Transport Barycenter, Ce Zhang, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Text documents are often interconnected in a network structure, e.g., academic papers via citations, Web pages via hyperlinks. On the one hand, though Graph Neural Networks (GNNs) have shown promising ability to derive effective embeddings for such networked documents, they do not assume a latent topic structure and result in uninterpretable embeddings. On the other hand, topic models can infer semantically interpretable topic distributions for documents by associating each topic with a group of understandable key words. However, most topic models mainly focus on plain text within documents and fail to leverage network structure across documents. Network connectivity reveals topic …

Mareva Injunction And Crypto Fraud, Man Yip

Research Collection Yong Pung How School Of Law

Without a Mareva injunction, there is always a risk that any victory on the substantive merits would be rendered hollow in practice by the absence of assets against which the judgment may be enforced. Indeed, it is in cases concerning fraud that the risk of a nugatory judgment is most amplified because the defendant would have the propensity to engage in dishonest practices. Alarmingly, nearly 30 years aft er Mercedez Benz, the ‘black hole’ has enlarged because fraudsters can easily hide behind complex corporate structures and the anonymity of cyberspace. 2 Th is chapter focuses on the use of Mareva …

Falsehoods, Foreign Interference, And Compelled Speech In Singapore, Kenny Chng

Research Collection Yong Pung How School Of Law

Online misinformation endangers the infrastructure of fact essential to public discourse and presents an even greater threat where it is being utilised as a weapon by hostile state actors. In recognition of these dangers, Singapore has implemented legal measures to combat online misinformation, enacting in quick succession the Protection from Online Falsehoods and Manipulation Act (POFMA) and the Foreign Interference (Countermeasures) Act (FICA). These statutes open up novel frontiers of development for Singapore's free speech jurisprudence. Indeed, these statutes confer upon government authorities the power to compel the authors of certain material to display notices stating that the material contains …

Negotiating A Brave New World: Fraud And Risk In Modern Commercial Law, Andrew B.L. Phang

Research Collection Yong Pung How School Of Law

We live in a brave new world. Much of what was once the stuff of fiction has now become reality. In particular, we have witnessed technological advances the like of which we would not even have begun to contemplate even at the turn of the second millennium. No one has been exempt – not even this mid-sexagenarian. Now, almost a quarter of a century on, we have to take stock of the double-edged sword that is technology. Indeed, technology has – unfortunately – facilitated fraud in a manner and on a scale that we have never witnessed before. With technology …

How Can Malaysian Courts Consistently Perform Meaningful Constitutional Rights Review? Lessons From Past Cases And The Way Forward, Benjamin Joshua Ong

Research Collection Yong Pung How School Of Law

In the past, Malaysian courts performing constitutional rights review played a merely clerical role, applying a test that was trivially easy for legislation to pass. Then a more rigorous proportionality test took root. However, the Federal Court in the 2020 case of Letitia Bosman whittled the test down again, and the courts once more played a minimal role in checking state action. The reasons for this cannot be explained merely by diversity in judicial philosophy or political contextual factors. Rather, the near-demise of proportionality (and, with it, robust constitutional review) was made possible by a lack of a clear sense …

Ten Years As Boundary Object: The Search For Identity And Belonging As 'Hongkongers', John Lowe, Darlene Machell Espena, George Wong

Research Collection College of Integrative Studies

This article examines the complex process of symbolic boundary-making of ‘Hongkonger’ cultural identities through the lens of the controversial 2015 film Ten Years, which is a celebrated omnibus production comprised of five short segments that picture a dystopic end to Hong Kong’s cherished way of life in the year 2025. The article is premised on an interdisciplinary approach engaging with cultural studies and film studies. On one hand, it explores how Ten Years functioned as a boundary object, a vast terrain within which cultural identities of what it means to be a Hongkonger are constructed, banished, imagined, and performed under …

Integration Of Urban Climate Research Within The Global Climate Change Discourse, Negin Nazarian, Benjamin Bechtel, Gerald Mills, Melissa Anne Hart, Ariane Middel, E. Scott Krayenhoff, Gaby S. Langendijk, Lei Zhao, Andy Pitman, Winston T. L. Chow

Research Collection College of Integrative Studies

Global climate science needs to address a fundamental challenge: the mismatch between the scales of anthropogenic processes driving change and the resulting climate impacts. While projected climate changes and impacts are global in extent, the drivers of this change, and the exposure to its impacts, are concentrated in densely populated urban areas. Despite occupying only 1–3% of the land, urban areas are home to most of the world’s population and responsible for ~70% of current greenhouse gas emissions [1]. By 2050, an additional 2.5 billion people are expected to live in urban areas, with up to 90% of this growth …

The Partnership Question As A Scale Question: Extending The Theorization Of Entrepreneurial Urbanism In China, Yong Zhang

Research Collection College of Integrative Studies

In this article I examine recent theoretical and empirical exchanges around partnership-based urban governance between North Atlantic and Chinese academics and policymakers. I argue that the latest wave of de jure private–public partnerships in urban China reflects an ongoing process of governance rescaling beyond conventional entrepreneurial urbanism theory. I propose an analytical framework that foregrounds successive experimental partnerships as tensions between institutional continuity and change arising from rescaling. In this study I examine variegated actually existing partnerships in Jiyuan, China, to identify generalizable ideal types of partnership-driven governance rescaling. I conclude by suggesting to enhance the theorization of entrepreneurial urbanism …

Vertical Expansion In The Making: Planning Against Deindustrialization By Promoting "Industry's Going Upstairs" In Shenzhen, Qianqian Wei, Yong Zhang

Research Collection College of Integrative Studies

In recent years, industrial metropolises in China have experienced a surge in proactive planning initiatives aimed at developing high-rise industrial structures, commonly known as “Industry’s Going Upstairs (IGU).” This study argues that the IGU represents a distinct form of urban verticality that is neither motivated by capitalist speculation nor sustainability prompts but rather by local states’ intervention to ensure economic resilience and enhance innovation capabilities. This study presents the case of Shenzhen to demonstrate how the adoption of the IGU initiative is catalyzed by concerns over manufacturing sectors’ out-migration and the effects of volatile US-China relations on the city’s competitiveness. …

Interpretable Tensor Fusion, Saurabh Varshneya, Antoine Ledent, Philipp Liznerski, Andriy Balinskyy, Purvanshi Mehta, Waleed Mustafa, Marius Kloft

Research Collection School Of Computing and Information Systems

Conventional machine learning methods are predominantly designed to predict outcomes based on a single data type. However, practical applications may encompass data of diverse types, such as text, images, and audio. We introduce interpretable tensor fusion (InTense), a multimodal learning method training a neural network to simultaneously learn multiple data representations and their interpretable fusion. InTense can separately capture both linear combinations and multiplicative interactions of the data types, thereby disentangling higher-order interactions from the individual effects of each modality. InTense provides interpretability out of the box by assigning relevance scores to modalities and their associations, respectively. The approach is …

Solving Long-Run Average Reward Robust Mdps Via Stochastic Games, Krishnendu Chatterjee, Ehsan Kafshdar Goharshady, Mehrdad Karrabi, Petr Novotný, Dorde Zikelic

Research Collection School Of Computing and Information Systems

Markov decision processes (MDPs) provide a standard framework for sequential decision making under uncertainty. However, MDPs do not take uncertainty in transition probabilities into account. Robust Markov decision processes (RMDPs) address this shortcoming of MDPs by assigning to each transition an uncertainty set rather than a single probability value. In this work, we consider polytopic RMDPs in which all uncertainty sets are polytopes and study the problem of solving long-run average reward polytopic RMDPs. We present a novel perspective on this problem and show that it can be reduced to solving long-run average reward turn-based stochastic games with finite state …

Task Scheduling Strategy For 3dpcp Considering Multidynamic Information Perturbation In Green Scene, Jianjia He, Jian Wu, Keng Siau

Research Collection School Of Computing and Information Systems

The 3D printing cloud platform (3DPCP) plays a pivotal role in breaking down the information silos between supply and demand, effectively reducing waste through information integration and intelligent production. However, due to the complexity of 3DPCP scheduling in green scenes and the multidynamic information perturbations, unveils problems in traditional task scheduling methods in 3DPCP. These issues manifest as incomplete considerations, subpar green performance, and weak adaptability to dynamic changes. There is an urgent need to design practical methods to realize the multidynamic information perturbations in green scenes within 3DPCP. Therefore, this article first defines the 3DPCP task scheduling problem for …

Peep With A Mirror: Breaking The Integrity Of Android App Sandboxing Via Unprivileged Cache Side Channel, Yan Lin, Joshua Wong, Xiang Li, Haoyu Ma, Debin Gao

Research Collection School Of Computing and Information Systems

Application sandboxing is a well-established security principle employed in the Android platform to safeguard sensitive information. However, hardware resources, specifically the CPU caches, are beyond the protection of this software-based mechanism, leaving room for potential side-channel attacks. Existing attacks against this particular weakness of app sandboxing mainly target shared components among apps, hence can only observe system-level program dynamics (such as UI tracing). In this work, we advance cache side-channel attacks by demonstrating the viability of non-intrusive and fine-grained probing across different app sandboxes, which have the potential to uncover app-specific and private program behaviors, thereby highlighting the importance of …