Knowledge tracing (KT) is a crucial technique to predict students' future performance by observing their historical learning processes. Due to the powerful representation ability of deep neural networks, remarkable progress has been made by using deep learning techniques to solve the KT problem. The majority of existing approaches rely on the \emph{homogeneous question} assumption that questions have equivalent contributions if they share the same set of knowledge components. Unfortunately, this assumption is inaccurate in real-world educational scenarios. Furthermore, it is very challenging to interpret the prediction results from the existing deep learning based KT models. Therefore, in this paper, we present QIKT, a question-centric interpretable KT model to address the above challenges. The proposed QIKT approach explicitly models students' knowledge state variations at a fine-grained level with question-sensitive cognitive representations that are jointly learned from a question-centric knowledge acquisition module and a question-centric problem solving module. Meanwhile, the QIKT utilizes an item response theory based prediction layer to generate interpretable prediction results. The proposed QIKT model is evaluated on three public real-world educational datasets. The results demonstrate that our approach is superior on the KT prediction task, and it outperforms a wide range of deep learning based KT models in terms of prediction accuracy with better model interpretability. To encourage reproducible results, we have provided all the datasets and code at \url{https://pykt.org/}. 

Improving Interpretability of Deep Sequential Knowledge Tracing Models with Dual Cognitive Representations

The networked-loan is major financing support for Micro, Small and Medium-sized Enterprises (MSMEs) in a number of countries. But external shocks may weaken the financial networks' robustness; an accidental default may spread across the network and collapse the whole network. Thus, predicting the critical firms in networked-loans to stem contagion risk and prevent potential systemic financial crises is of crucial significance to the long-term health of inclusive finance and sustainable economic development. Existing approaches in the banking industry dismiss the contagion risk across loan networks and need extensive knowledge with sophisticated financial expertise. Regarding the issues, we propose a novel approach to predict critical firms for stemming contagion risk in the bank industry with deep reinforcement learning integrated with high-order graph message-passing networks. We demonstrate that our approach outperforms the state-of-the-art baselines significantly on the dataset from a large commercial bank. Moreover, we also conducted empirical studies on the real-world loan dataset for risk mitigation. The proposed approach enables financial regulators and risk managers to better track and understands contagion and systemic risk in networked-loans. The superior performance also represents a paradigm shift in addressing the modern challenges in financing support of MSMEs and sustainable economic development.

Critical Firms Prediction for Stemming Contagion Risk in Networked-loans through Graph-based Deep Reinforcement Learning

Despite improvements in safe water and sanitation services in low-income countries, a substantial proportion of the population in Africa still does not have access to these essential services. Up-to-date fine-scale maps of low-income settlements are urgently needed by authorities  to improve service provision. We aim to develop a cost-effective solution to generate fine-scale maps of these vulnerable populations using multi-source public information. The problem is challenging as ground-truth maps are available at only a limited number of cities, and the patterns are heterogeneous across cities. Recent attempts tackling the spatial heterogeneity issue focus on scenarios where true labels partially exist for each input region, which are unavailable for the present problem. We propose a dynamic point-to-region co-learning framework to learn heterogeneity patterns that cannot be reflected by point-level information and generalize deep learners to new areas with no labels. We also propose an attention-based correction layer to remove spurious signatures, and a region-gate to capture both region-invariant and variant patterns. Experiment results on real-world fine-scale data in three cities of Kenya show that the proposed approach can largely improve model performance on various base network architectures.

Point-to-Region Co-Learning for Poverty Mapping at High Resolution Using Satellite Imagery

The popularity of on-demand ride pooling is owing to the benefits offered to customers (lower prices), taxi drivers (higher revenue), environment (lower carbon footprint due to fewer vehicles) and aggregation companies like Uber (higher revenue). To achieve these benefits, two key interlinked challenges have to be solved effectively: (a) pricing -- setting prices to customer requests for taxis; and (b) matching -- assignment of customers (that accepted the prices) to taxis/cars. Traditionally, both these challenges have been studied individually and using myopic approaches (considering only current requests), without considering the impact of current matching on addressing future requests. In this paper, we develop a novel framework that handles the pricing and matching problems together, while also considering the future impact of the pricing and matching decisions. In our experimental results on a real-world taxi dataset, we demonstrate that our framework can significantly improve revenue (up to 17\% and on average 6.4\%) in a sustainable manner by reducing the number of vehicles  (up to 14\% and on average 10.6\%) required to obtain a given fixed revenue and the overall distance travelled by vehicles (up to 11.1\% and on average 3.7\%). That is to say, we are able to provide an ideal win-win scenario for all stakeholders (customers, drivers, aggregator, environment) involved by obtaining higher revenue for customers, drivers, aggregator (ride pooling company) while being good for the environment (due to fewer number of vehicles on the road and lesser fuel consumed).

Future Aware Pricing and Matching for Sustainable On-demand Ride Pooling

The mobile communication enabled by cellular networks is the one of the main foundations of our modern society. Optimizing the performance of cellular networks and providing massive connectivity with improved coverage and user experience has a considerable social and economic impact on our daily life. This performance relies heavily on the configuration of the network parameters. However, with the massive increase in both the size and complexity of cellular networks, network management, especially parameter configuration, is becoming complicated. The current practice, which relies largely on experts' prior knowledge, is not adequate and will require lots of domain experts and high maintenance costs. In this work, we propose a learning-based framework for handover parameter configuration. The key challenge, in this case, is to tackle the complicated dependencies between neighboring cells and jointly optimize the whole network. Our framework addresses this challenge in two ways. First, we introduce a novel approach to imitate how the network responds to different network states and parameter values, called auto-grouping graph convolutional network (AG-GCN). During the parameter configuration stage, instead of solving the global optimization problem, we design a local multi-objective optimization strategy where each cell considers several local performance metrics to balance its own performance and its neighbors. We evaluate our proposed algorithm via a simulator constructed using real network data. We demonstrate that the handover parameters our model can find, achieve better average network throughput compared to those recommended by experts as well as alternative baselines, which can bring better network quality and stability. It has the potential to massively reduce costs arising from human expert intervention and maintenance.

Neighbor Auto-Grouping Graph Neural Networks for Handover Parameter Configuration in Cellular Network

Generating human mobility trajectories is of great importance to solve the lack of large-scale trajectory data in numerous applications, which is caused by privacy concerns. However, existing mobility trajectory generation methods still require real-world human trajectories centrally collected as the training data, where there exists an inescapable risk of privacy leakage. To overcome this limitation, in this paper, we propose PateGail, a privacy-preserving imitation learning model to generate mobility trajectories, which utilizes the powerful generative adversary imitation learning (GAIL) model to simulate the decision-making process of humans. Further, in order to protect user privacy, we train this model collectively based on decentralized mobility data stored in user devices, where personal discriminators are trained locally to distinguish and reward the real and generated human trajectories. In the training process, only the generated trajectories and their rewards obtained based on personal discriminators are shared between the server and devices, whose privacy is further preserved by our proposed perturbation mechanisms with theoretical proof to satisfy differential privacy. Further, to better model the human decision-making process, we propose a novel aggregation mechanism of the rewards obtained from personal discriminators. We theoretically prove that under the reward obtained based on the aggregation mechanism, our proposed model maximizes the lower bound of the discounted total rewards of users. Extensive experiments show that the trajectories generated by our model are able to resemble real-world trajectories in terms of five key statistical metrics, outperforming state-of-the-art algorithms by over 38.19%. Furthermore, we demonstrate that the synthetic trajectories are able to efficiently support practical applications, including mobility prediction and location recommendation.

PateGail: A Privacy-preserving Mobility Trajectory Generator with Imitation Learning

The concept of walkable urban development has gained increased
attention due to its public health, economic, and environmental
sustainability benefits. Unfortunately, land zoning
and historic under-investment have resulted in spatial inequality
in walkability and social inequality among residents.
We tackle the problem of Walkability Optimization through
the lens of combinatorial optimization. The task is to select
locations in which additional amenities (e.g., grocery stores,
schools, restaurants) can be allocated to improve resident access
via walking while taking into account existing amenities
and providing multiple options (e.g., for restaurants).
To this end, we derive Mixed-Integer Linear Programming
(MILP) and Constraint Programming (CP) models. Moreover,
we show that the problem’s objective function is submodular
in special cases, which motivates an efficient greedy
heuristic. We conduct a case study on 31 underserved neighborhoods
in the City of Toronto, Canada. MILP finds the
best solutions in most scenarios but does not scale well with
network size. The greedy algorithm scales well and finds
near-optimal solutions. Our empirical evaluation shows that
neighbourhoods with low walkability have a great potential
for transformation into pedestrian-friendly neighbourhoods
by strategically placing new amenities. Allocating 3 additional
grocery stores, schools, and restaurants can improve the
“WalkScore” by more than 50 points (on a scale of 100) for 4
neighbourhoods and reduce the walking distances to amenities
for 75% of all residential locations to 10 minutes for all
amenity types. Our code and paper appendix are available at
https://github.com/khalil-research/walkability.

Walkability Optimization: Formulations, Algorithms, and a Case Study of Toronto

Human migration into urban areas has increased urbanization worldwide. As natural areas become urbanized, animals are forced to migrate, adapt, or perish in response to drastic habitat changes. Studies have documented how animals respond to the changes in biotic and abiotic factors associated with urbanization. Mostly, these studies document changes in phenotypic traits—morphology, physiology, and/or behaviors. With so many studies now available, meta-analyses offer insights into whether general trends in phenotypic responses among animals occur with urbanization. My lab has used phylogenetic meta-analyses to understand whether certain animal groups exhibit consistent phenotypic responses to urbanization, and to determine if the phenotypes of animals are consistently different across cities within studies. We have found that lizards—an animal group which has many urban-dwelling species—generally respond to urbanization through changes in body size. Urban lizards are larger than their non-urban counterparts. However, among all animal groups, the effects of urbanization depend on how many cities are included in a study, and the geographic distance between those cities. Phenotypic differences between urban populations of the same species increase as the more cities are investigated and as the farther away cities are from each other. Thus, separate urban populations of the same species can diverge phenotypically. Overall, it remains difficult to generalize the effects of urbanization on animal phenotypes with the current state of the literature. As more research emerges, meta-analyses will help us recognize if and how animals respond to human-induced habitat changes.

Using meta-analyses to understand phenotypic responses to urbanization in animals

Multinational corporations and states are driving a process of commodifying and categorizing land. At the same time, the material and social realities of landscapes resist this cooptation. Our panel explores the frictions between different visions and values of land. This panel focuses on the technocrats and bureaucrats who mediate environmental and legal frictions over land. We draw on a range of sources • Indigenous land rights cases, forestry concessions, land remediation plans, and urban planning agendas • to explore how these mediators (re)produce the meaning of land. We also explore why frictions remain despite these policies and procedures. Environmental remediation planners use predictive models, but the realities of ruined landscapes escape those carefully crafted plans. Legal mediators attempt to demarcate Indigenous lands from extractive concessions, but those boundaries inevitably overlap. City planners designate “illegal” vs” legal” urban settlements, but the people who live in those settlements resist those designations. We look at how material realities and local actors disrupt the commodification and categorization of land. We draw on historical archives and ethnography to explore how technology, planning, and law mediate these frictions. What are the roles of local power structures, the state, and multinational corporations in determining land’s use and value? What are emerging counter-narratives? How do these narratives develop and become inscribed on the land itself? Ultimately, we ask what these histories and contemporary realities tell us about the persistence of frictions over land and how those frictions will define land in the future.

Grounded Frictions: Land re(mediation) and (re)production

Planning, Mapping, and Resisting Panama's Wild Darién

What forms of urban politics come into relief if we foreground cities as sites of excessive relationalities, rather than scarce rights? How might an ethics of mutuality prefigure the demand for rights in urban polities? Since the 1970s, approaches to social and economic justice in urban theory have coalesced around the theme of “the right to the city” (Harvey 2009; 2013; Lefebvre 1996). The problematic of rights brings into view important struggles to organize and protect public space (Low 2000; Peterson 2006), practices of appropriating street and market sites in order to make a living in the face of exclusionary policies (Bayat 2012; Bhimji 2010; Grant and Thompson 2015), and the more general possibilities of civic engagement and protest in cities as key sites in today’s financialized global economy (Leitner and Strunk 2014). Yet even as urban inhabitants mobilize a universalist discourse of “rights,” their understandings and enactments come nested in relations that far exceed abstractions of civil society and law (Chatterjee 2003, Simone 2004), relations that are situated in infrastructures and ecologies that accrue and retrace sedimented histories of belonging, othering, and coloniality (Anand 2017; Caldeira 2000; Escobar 2004; Quijano and Ennis 2000). This panel grapples with two problematics we read in the influential concept of rights to the city. The first problematic is the idea that social justice is universal, rational, and depersonalized•and that in cities, it necessarily involves access to public space and equality of urban residents. In practice, we observe that, even while claiming rights, urban residents claim and control urban space and resources through practices that are personalized, embodied, and networked through caste, ethnic, racial, and gendered geographies (Agard Jones 2013; Jamaal Wright 2020; Lee 2017; Roberts 2016; Sreenath 2019); exceeding conceits of modernist rationality and to demarcate forms of kin and community making that, far from universal, are situated in ecologies that register as Western universalism’s ‘imperial debris’ (Stoller 2013; Tsing 2015). The second problematic, which we understand as intimately related to the first, is the way ontologies of scarcity continue to underpin the notion of urban rights, even as urban scarcities come circumscribed in unruly relationalities and toxicities (Chalfin 2017; Fredricks 2018). In such framings, urban space, citizenship, and resources appear in terms of their lack: scarce resources to be distributed among coherent and pre-constituted publics. The panel invites thinking about other political ontologies emergent in urban spaces (de la Cadena 2014; Massey 2005; Tedesco 2014). It opens dialogue on how problematizing scarcity as a social construct invites other understandings of social justice that account for economies of excess, practices of mutuality, enactments of care and speculation, and other means of distributing resources and claims to resources (such as credit and debt) in cities today.

Underline digital video library

Improving Interpretability of Deep Sequential Knowledge Tracing Models with Dual Cognitive Representations

Critical Firms Prediction for Stemming Contagion Risk in Networked-loans through Graph-based Deep Reinforcement Learning

Point-to-Region Co-Learning for Poverty Mapping at High Resolution Using Satellite Imagery

Future Aware Pricing and Matching for Sustainable On-demand Ride Pooling

Neighbor Auto-Grouping Graph Neural Networks for Handover Parameter Configuration in Cellular Network

PateGail: A Privacy-preserving Mobility Trajectory Generator with Imitation Learning

Walkability Optimization: Formulations, Algorithms, and a Case Study of Toronto

Using meta-analyses to understand phenotypic responses to urbanization in animals

Grounded Frictions: Land re(mediation) and (re)production

Planning, Mapping, and Resisting Panama's Wild Darién

Missing Money, Excessive time, and Deferred Responsibilities in an Ethiopian Frontier City

Toxic possessions: Some practicalities of dirt from Detroit

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES