Opening Address by Dr. Janil Puthucheary, Senior Minister of State, Ministry of Communications and Information & Ministry of Health for Knowledge Discovery and Data Mining (KDD) Conference 2021

Back to Speeches

Opening Address for Knowledge Discovery and Data Mining (KDD) Conference 2021 by Dr. Janil Puthucheary, Senior Minister of State, Ministry of Communications and Information & Ministry of Health

 

Introduction

1. Good morning! I am delighted to join everyone at KDD 2021. Special congratulations to the Organising Committee as well for putting this together in such challenging times. I would also like to thank Temasek Holdings for bringing KDD 2021 to Singapore, and for inviting me to participate alongside global experts in Data Science and AI.

2. These two technologies, Data Science and AI, are important for Singapore’s Smart Nation initiative. We launched our National AI Strategy in 2019 in recognition of the potential for AI to fundamentally transform the way we live, work and play. Our vision is to be a leader in developing and deploying scalable, impactful AI solutions, in key sectors of high value and relevance to our citizens and businesses by 2030.

3. Several enablers need to be in place for us to achieve this. One is the presence of a vibrant Research, Innovation and Enterprise, or RIE, ecosystem for the research and development of Data Science and AI capabilities. Today, if I may touch on three areas that Singapore is working on, to build up our Data Science and AI RIE ecosystem. They are:

a. Firstly, supporting cutting-edge research;

b. Secondly, promoting data sharing; and

c. Finally, translating that research into real-world applications, possibly the most important aspect.

Supporting cutting-edge research

4. First, we are committed to investing in cutting-edge research.

5. As part of Singapore’s $25 billion five-year RIE 2025 plan, we will step up our investments in AI and data, to strengthen our digital capabilities, focused on addressing the needs of our economy and society.

6. For example, we have invested $300 million to set up AI Singapore, a national R&D programme. Recently, we committed a further $120 million to this to be used over the next five years. AI Singapore supports activities across the spectrum of:

a. fundamental research, in areas such as explainable AI and resource-efficient AI;

b. translational research to address national challenges such as healthcare; and

c. industry collaborations to accelerate AI adoption.

7. We are also setting up new research centres.

a. We have recently set up two that focus on privacy-preserving technologies. One at the National University of Singapore and the other at Nanyang Technological University.

b. This is recognising the need to engender trust in the digital world. Businesses and consumers need to feel safe using digital technologies, and be assured that their privacy is protected. By enhancing our research capabilities in privacy-preserving technologies, we hope to provide all stakeholders with that assurance and confidence as they digitalise, and give them the confidence to unlock new opportunities across the economy.

c. It is great that one of the focuses for KDD Trust Day is on privacy-preserving AI technologies. We look forward to the exchange of ideas, sharing of best practices, and forging of new partnerships in this relatively new area.

8. I am glad that we are making good progress in developing a vibrant RIE ecosystem.

a. Singapore is currently ranked first globally in terms of the Field-Weighted Citation Impact (FWCI) of our AI publications; and

b. Two out of ten of the world’s most promising young AI researchers ranked by the Institute of Electrical and Electronics Engineers (IEEE) in 2020 are based in Singapore.

9. Going forward, we hope that the research landscape in Singapore will continue to present many exciting opportunities, including collaborations with international partners and friends like yourselves.

Promoting data sharing

10. Beyond investing in cutting-edge research, we are also committed to promoting data sharing.

11. Digital transactions are becoming increasingly complicated and complex, involving data from multiple sources and then interfacing with multiple parties. With increased data sharing, we will be able to construct richer datasets, which in turn drives more efficient, robust, and innovative AI solutions.

12. There are, however, a number of challenges – a lack of common data standards, concerns over data privacy, and suboptimal data infrastructure.

13. Singapore hopes to overcome some of these challenges by developing new data sharing infrastructure that is open, accessible, and secure. We are taking a sectoral approach, as each sector and industry has specific needs and different competitive landscapes. We are bringing together stakeholders within each sector to identify key data assets and to develop central technical platforms to support such data sharing. There has been progress.

a. For instance, in the transport sector, we have developed the Land Transport DataMall, which publishes more than 100 different types of transport-related datasets. For example, real-time bus transport patterns, taxi availability, and traffic conditions. Today, there are more than 600 million data queries processed through this every month, supporting novel AI use cases.

                     i. Most recently, researchers from the Singapore University of Technology and Design (SUTD) used millions of origin-destination bus and train ridership datapoints from the DataMall to pinpoint potential Covid-19 super-spreader locations, so as to help stem the spread of the disease in Singapore.

b. Beyond transportation, we are also developing data platforms in other sectors. Last month, for example, we announced the launch of a common data infrastructure for the supply chain ecosystem, called the Singapore Trade Data Exchange, or SGTraDex.

                     i. SGTraDex provides stakeholders with a ‘plug and play’ platform to exchange data in a secure environment.  Data is encrypted and transmitted without being stored. From the early results of pilots so far, we see that SGTraDex has the potential to unlock more than $200 million of value annually for the supply chain ecosystem.

14. Apart from developing new data infrastructure, we are also working with credible and reliable partners internationally to facilitate trusted and secure cross-border data flows.

a. At the regional level, we are spearheading efforts such as the ASEAN Cross Border Data Flows Mechanism.

b. At the international level, we have concluded two Digital Economy Agreements – one with Chile and New Zealand, and one with Australia. And there are several other agreements in the pipeline. 

c. We look forward to forging even more regional and international partnerships to facilitate greater data-sharing in the months and years to come.

Translating research into real-world applications

15. Finally, I would like to share about our efforts to translate all these research into real-world applications.

16. AI holds great promise as the next frontier of technological opportunities. The readiness of the technology, and the challenges of deployment, however, sometimes slow down AI adoption.

17. We are learning to overcome such challenges as we implement our National AI Projects. We have launched an initial tranche of five National AI Projects, in sectors such as (1) transport and logistics, (2) smart cities and estates, (3) healthcare, (4) education, and (5) safety and security.

a. For example, a key use case under our National AI Project in healthcare is SELENA+, a locally developed AI algorithm that analyses eye scans to detect diabetic retinopathy. SELENA+ achieves accuracy levels of more than 90% and provides an analysis of these scans in mere seconds.

b. There have been challenges despite this model’s performance that had to be resolved, in order for the deployment in real-world settings. Teams had to come together. Healthcare practitioners, healthcare regulators, and our private sector integrators had to work with public health institutions to address a range of deployment issues -  regulatory compliance, funding challenges, models of clinical service.

c. But today, SELENA+ is projected to screen 120,000 patients annually, covering all diabetic eye screenings at our public health primary care centres across Singapore. It is excellent to see this kind of progress we are making.

18. In addition to our National AI Projects, we have set up testbed sites to trial new AI innovations, which allows us to bring that research from the lab into a variety of real-world settings. For example:

a. The Maritime Drone Estate (MDE), is a testbed for drone technologies like shore-to-ship deliveries and remote ship inspections;

b. We have the Sustainable Tropical Data Centre Testbed (STDCT) to trial cooling technologies like digital twins and energy optimisation algorithms for data centres central to our Smart Nation plans; and

c. We have the Centre for Healthcare Innovation Start-up Enterprise Link (CHISEL), to testbed innovative healthcare technologies like medical imaging and clinical risk scoring solutions and think how to translate these into business models.

19. One particular testbed site that I would like to elaborate on is the Punggol Digital District, or PDD. PDD is a 50-hectare greenfield site that serves as a living lab to test how digital technologies can enable new concepts of living, working, and service delivery. Opportunities to build an entire digital district from scratch do not come by every day. Changes to how cities are run are often incremental, and rarely revolutionary. PDD therefore offers us a unique opportunity to shift the paradigm in the way technology is used to manage our infrastructure, and provide services to citizens and businesses. We already have numerous initiatives planned for PDD. These include:

a. A smart transport system that involves the pilot deployment of autonomous buses

b. A green and sustainable estate that is supported by AI-enabled home appliances, intelligent estate lighting infrastructure, and a smart energy grid full of solutions to optimise energy consumption and reduce carbon emissions; and

c. New operating and service delivery concepts, such as

                    i. food centres with smart facilities management systems to provide better cleaning and maintenance services, and

                    ii. sports centres with virtual coaching, augmented reality gamification and online facilities booking.

20. There are many opportunities yet to be explored, whether through our National AI Projects or the many testbed sites that we are building in Singapore. I am thrilled to see how these initiatives can potentially unlock new innovative solutions from AI and improve the lived experiences of our residents in Singapore.

Conclusion

21. AI is a transformative technology, and I have given you a glimpse of how Singapore is approaching this innovation and technology. There is a lot more we can and must do. We welcome partners from the global community, to exchange insights and experiences, and collaborate with us on our journey to become a Smart Nation.

22. I wish everyone good health and a useful, productive, and enriching conference ahead. Thank you very much.

Last updated on 15 Sep 2021