Portrait of Shin (Alexandre) Koseki

Shin (Alexandre) Koseki

Affiliate Member
Assistant Professor, Université de Montréal, School of Urban Planning and Landscape Architecture
Research Topics
Data Mining
Website
LinkedIn
X

Biography

Shin Koseki is an assistant professor at the School of Urban Planning and Landscape Architecture of the Faculty of Environmental Design, Université de Montréal. He is also the director and chairholder of the UNESCO Chair in Urban Landscape. Trained in architecture and urban planning in Canada and Switzerland, Koseki is interested in the integration of new technologies in planning practices, the contribution of interactive democracy to the sustainable development of territories, and the role of public space in the acquisition of knowledge and skills. His research interests include the application of AI systems in urban design and new processes of environmental and technological governance.

In 2022, Koseki co-authored the white paper produced jointly by Mila – Quebec Artificial Intelligence Institute and UN-Habitat entitled “AI & Cities: Risks, Applications and Governance.”

His research on codesigning responsible AI systems in cities is supported by the New Frontiers in Research Fund and the Quebec Ministry of Economy, Innovation and Energy. Koseki has conducted research at the Swiss Federal Institute of Technology (Lausanne and Zurich), University of Oxford, National University of Singapore, Massachusetts Institute of Technology (MIT), University of Zurich, and Max Planck Institute for Art and Architecture History (Bibliotheca Hertziana). Back in his home city of Montréal, he works with his students on projects to revitalize and renaturalize the St. Lawrence River, as well as improve the quality of life of communities living along its shores.

Current Students

Hugo Berard
Postdoctorate - Université de Montréal
Leandry Junior JIEUTSA NKUIDJE Jieutsa
PhD - Université de Montréal
Github
Rashid Mushkani
PhD - Université de Montréal
Website
Github
Google Scholar
Frédérique Roy
Master's Research - Université de Montréal

Publications

Negotiative Alignment: Embracing Disagreement to Achieve Fairer Outcomes -- Insights from Urban Studies
Rashid A. Mushkani
Hugo Berard
Shin (Alexandre) Koseki
2025-03-16
ArXiv (preprint)
arxiv.org
arxiv.org
LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spaces
Rashid A. Mushkani
Shravan Nayak
Hugo Berard
Allison Cohen
Shin (Alexandre) Koseki
Hadrien Bertrand
2025-02-27
ArXiv (preprint)
arxiv.org
arxiv.org
LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spaces
Rashid A. Mushkani
Shravan Nayak
Hugo Berard
Allison Cohen
Shin (Alexandre) Koseki
Hadrien Bertrand
We introduce the Local Intersectional Visual Spaces (LIVS) dataset, a benchmark for multi-criteria alignment of text-to-image (T2I) models i… (see more)n inclusive urban planning. Developed through a two-year participatory process with 30 community organizations, LIVS encodes diverse spatial preferences across 634 initial concepts, consolidated into six core criteria: Accessibility, Safety, Comfort, Invitingness, Inclusivity, and Diversity, through 37,710 pairwise comparisons. Using Direct Preference Optimization (DPO) to fine-tune Stable Diffusion XL, we observed a measurable increase in alignment with community preferences, though a significant proportion of neutral ratings highlights the complexity of modeling intersectional needs. Additionally, as annotation volume increases, accuracy shifts further toward the DPO-tuned model, suggesting that larger-scale preference data enhances fine-tuning effectiveness. LIVS underscores the necessity of integrating context-specific, stakeholder-driven criteria into generative modeling and provides a resource for evaluating AI alignment methodologies across diverse socio-spatial contexts.
2025-02-27
ArXiv (preprint)
arxiv.org
arxiv.org
AI-EDI-SPACE: A Co-designed Dataset for Evaluating the Quality of Public Spaces
S. Gowaikar
Hugo Berard
Rashid A. Mushkani
Emmanuel Beaudry Marchand
Toumadher Ammar
Shin (Alexandre) Koseki
Advancements in AI heavily rely on large-scale datasets meticulously curated and annotated for training. However, concerns persist regarding… (see more) the transparency and context of data collection methodologies, especially when sourced through crowdsourcing platforms. Crowdsourcing often employs low-wage workers with poor working conditions and lacks consideration for the representativeness of annotators, leading to algorithms that fail to represent diverse views and perpetuate biases against certain groups. To address these limitations, we propose a methodology involving a co-design model that actively engages stakeholders at key stages, integrating principles of Equity, Diversity, and Inclusion (EDI) to ensure diverse viewpoints. We apply this methodology to develop a dataset and AI model for evaluating public space quality using street view images, demonstrating its effectiveness in capturing diverse perspectives and fostering higher-quality data.
2024-11-01
arXiv (published)
doi.org
arxiv.org
AI-EDI-SPACE: A Co-designed Dataset for Evaluating the Quality of Public Spaces
S. Gowaikar
Hugo Berard
Rashid A. Mushkani
Emmanuel Beaudry Marchand
Toumadher Ammar
Shin (Alexandre) Koseki
Advancements in AI heavily rely on large-scale datasets meticulously curated and annotated for training. However, concerns persist regarding… (see more) the transparency and context of data collection methodologies, especially when sourced through crowdsourcing platforms. Crowdsourcing often employs low-wage workers with poor working conditions and lacks consideration for the representativeness of annotators, leading to algorithms that fail to represent diverse views and perpetuate biases against certain groups. To address these limitations, we propose a methodology involving a co-design model that actively engages stakeholders at key stages, integrating principles of Equity, Diversity, and Inclusion (EDI) to ensure diverse viewpoints. We apply this methodology to develop a dataset and AI model for evaluating public space quality using street view images, demonstrating its effectiveness in capturing diverse perspectives and fostering higher-quality data.
2024-11-01
ArXiv (preprint)
doi.org
arxiv.org
From Efficiency to Equity: Measuring Fairness in Preference Learning
S. Gowaikar
Hugo Berard
Rashid A. Mushkani
Shin (Alexandre) Koseki
As AI systems, particularly generative models, increasingly influence decision-making, ensuring that they are able to fairly represent diver… (see more)se human preferences becomes crucial. This paper introduces a novel framework for evaluating epistemic fairness in preference learning models inspired by economic theories of inequality and Rawlsian justice. We propose metrics adapted from the Gini Coefficient, Atkinson Index, and Kuznets Ratio to quantify fairness in these models. We validate our approach using two datasets: a custom visual preference dataset (AI-EDI-Space) and the Jester Jokes dataset. Our analysis reveals variations in model performance across users, highlighting potential epistemic injustices. We explore pre-processing and in-processing techniques to mitigate these inequalities, demonstrating a complex relationship between model efficiency and fairness. This work contributes to AI ethics by providing a framework for evaluating and improving epistemic fairness in preference learning models, offering insights for developing more inclusive AI systems in contexts where diverse human preferences are crucial.
2024-10-24
ArXiv (preprint)
doi.org
arxiv.org
From Efficiency to Equity: Measuring Fairness in Preference Learning
S. Gowaikar
Hugo Berard
Rashid A. Mushkani
Shin (Alexandre) Koseki
As AI systems, particularly generative models, increasingly influence decision-making, ensuring that they are able to fairly represent diver… (see more)se human preferences becomes crucial. This paper introduces a novel framework for evaluating epistemic fairness in preference learning models inspired by economic theories of inequality and Rawlsian justice. We propose metrics adapted from the Gini Coefficient, Atkinson Index, and Kuznets Ratio to quantify fairness in these models. We validate our approach using two datasets: a custom visual preference dataset (AI-EDI-Space) and the Jester Jokes dataset. Our analysis reveals variations in model performance across users, highlighting potential epistemic injustices. We explore pre-processing and in-processing techniques to mitigate these inequalities, demonstrating a complex relationship between model efficiency and fairness. This work contributes to AI ethics by providing a framework for evaluating and improving epistemic fairness in preference learning models, offering insights for developing more inclusive AI systems in contexts where diverse human preferences are crucial.
2024-10-24
ArXiv (preprint)
doi.org
arxiv.org
Deployment of digital technologies in African cities: emerging issues and policy recommendations for local governments
Leandry Jieutsa
Irina Gbaguidi
Wijdane Nadifi
Shin (Alexandre) Koseki
2024-04-01
Data & Policy (published)
doi.org
Evaluation algorithmique inclusive de la qualité des espaces publics
Shin (Alexandre) Koseki
Toumadher Ammar
Rashid Ahmad Mushkani
Hugo Berard
Sarah Tannir
2024-01-01
SHS Web of Conferences (published)
doi.org