��ݮ��Ƶ News - Math

Firebird forecasts and models wildfire spread

Angelica Sanchez — Wed, 20 Aug 2025 12:00:00 +0000

From co-op to startup: ��ݮ��Ƶ students develop an AI platform that uses real-time data for wildfire prediction and early prevention

As wildfires grow more frequent and severe across Canada, University of ��ݮ��Ƶ researchers and students are stepping up with innovative, data-driven solutions.

Among them is Faculty of Science student Dante Maggiotto, and his team of student researchers who are developing Firebird — a cutting-edge platform that integrates artificial intelligence (AI), satellite imagery and meteorological data to forecast high-risk zones by modeling wildfires and how they spread.

“Now is the most important time to be developing solutions for climate-related challenges,” Maggiotto says. “We want to move wildfire management from reactive firefighting to proactive mitigation.”

The urgency is clear: around 60 per cent of all land burned in Canada due to wildfires over the past decade occurred during just the 2023 and 2024 seasons, Maggiotto explains. The 2016 Fort McMurray fires alone caused nearly $10 billion in damages, making them the most economically devastating natural disaster in Canadian history.

Firebird is one of many ventures emerging from ��ݮ��Ƶ’s innovation ecosystem, where students are empowered to tackle global challenges through interdisciplinary research, entrepreneurial thinking and a commitment to building a sustainable future.

With support from programs like Enterprise Co-op (E Co-op) and Velocity, Maggiotto and his team are turning their research in wildfire forecasting and prevention into scalable, data-driven solutions.

Firebird's early prototype of an Alberta grid visualizing wildfire ignition and spread.

From reactive firefighting to proactive mitigation

In wildfire management, every hour counts — the difference between a contained spark and a regional disaster can be measured in minutes. Firebird’s technology analyzes spatial-temporal patterns to generate predictive maps that pinpoint high-risk areas with precision, before fires ignite. This enables governments, industries and fire crews to allocate resources more effectively and respond before small blazes escalate into uncontrollable disasters.

Unlike traditional active wildfire models, Firebird uses deep learning to process large volumes of data in real time. Firebird’s hybrid approach of combining satellite and ground-based sensing allows for faster, more accurate forecasting that’s continuously validated against actual fire incidents.

Firebird combines satellite and ground-based sensing to visualize wildfire spread.

“Canada is in the midst of its second-worst wildfire season on record, with more than 7.3 million hectares burned as of mid-August — an area the size of New Brunswick,” explains Ibraheem Azhar, co-founder of Firebird and a fourth-year Science and Business student. 

“That’s more than double the 10-year average. With three [wildfire seasons] in the last three years now ranking among the worst in Canadian history, it’s clear these wildfire seasons aren’t anomalies — they’re becoming the new normal.”

The team is also developing an industry-grade wildfire spread model that applies physics-based principles — such as fluid and combustion dynamics — to forecast how fires move across different terrains. Running on high-performance computing, the system integrates imagery, human activity, weather and topographic data to deliver scalable, real-time simulations.

Firebird's industry-grade wildfire spread model that applies physics-based principles — such as fluid and combustion dynamics — to forecast how fires move across different terrains

“Our mission is to lead the charge in environmental stewardship by deeply understanding wildfires,” Maggiotto says. “We want Firebird to be the go-to tool in wildfire resilience — protecting forests, safeguarding communities and empowering fire crews who defend them.”

Through its Application Programming Interface (API) and Software as a Service (SaaS) licensing model, Firebird allows government agencies to automatically receive wildfire risk updates and integrate them into their own systems. The forecasts are easy to visualize through interactive map layers and accessible via a user-friendly online platform.

Firebird is currently modeling across Alberta and plans to begin pilot testing in 2026, supported by early impact case studies.

From research idea to impact: Powered by campus supports

Maggiotto, a fourth-year co-op student, co-founded Firebird alongside fellow students Ibraheem Azhar and Teodora Vujovic, as well as alum J. Romero (BSc ’24). As Albertan residents, Azhar and Romero felt a strong connection to the issue and knew that building a scalable solution would require an interdisciplinary approach. Together, the team brings expertise in computer science, business, fire science and forest management— uniting their strengths to address one of Canada’s most pressing challenges.

Many students join the E Co-op program to launch their own businesses while earning a standard 16-week co-op credit. For Maggiotto, the program offered more than just academic credit — it was a launchpad for Firebird.

With guidance from Dr. Wayne Chang, E Co-op coordinator, Maggiotto received valuable feedback and networking opportunities that helped transform Firebird from a student project into a scalable startup.

“Dr. Chang has an incredible network within ��ݮ��Ƶ’s ecosystem and beyond,” Maggiotto says. “He made key introductions that helped us connect with industry experts and gain early traction. Those connections were instrumental in accelerating Firebird’s development far beyond what we could have imagined.”

As part of the program, Maggiotto pitched Firebird at the Spring 2025 E Co-op Pitch Competition, where he secured $5,000 in funding through the E-Launch Award — Firebird’s first external funding.

During the same term, the Firebird team joined Velocity’s inaugural Summer Accelerator — an intensive program hosted at the Innovation Arena to support student founders. Around 20 startup teams received mentorship, legal and business guidance, and introductions to early customers.

“Demo Day was a pinnacle moment,” Maggiotto explains. “It gave us a chance to reflect on our progress and share our vision for what’s next. But the most impactful part was working in a community of founders who inspired and motivated us daily."

Maggiotto’s entrepreneurial drive was also shaped by years of involvement with Socratica — one of Canada's fastest-growing student maker collectives — and a range of student-led organizations at ��ݮ��Ƶ.

Through programs like E Co-op and Velocity, ��ݮ��Ƶ empowers students to turn bold ideas into real-world solutions. With access to mentorship, funding and a vibrant community of innovators, students like Maggiotto are equipped to navigate the startup journey with confidence and transform classroom concepts into real-world ventures.

Technological Futures

University of ��ݮ��Ƶ launches chair in physics of information and AI

Angelica Sanchez — Wed, 06 Aug 2025 20:09:23 +0000

The Dieter Schwarz Foundation’s transformational gift will support a global research hub in exploratory science and technology

New insight into the laws of nature and advances in quantum technology may soon be possible thanks to a transformational gift from the Dieter Schwarz Foundation.

With its first investment in a Canadian university, the prominent German foundation has established the Dieter Schwarz Chair in the Physics of Information and AI at the University of ��ݮ��Ƶ, with an aim to accelerate research with exciting possibilities for industry and society.

The physics of information provides a bridge between quantum theory and advanced technologies, including quantum computing and communication. With the rise of artificial intelligence, the field has expanded to include machine learning and quantum algorithm design. Discoveries in this growing discipline promise to inspire innovation in areas ranging from medicine to materials science to cybersecurity.

Housed in the Faculty of Mathematics, the Dieter Schwarz Chair will be held by Dr. Achim Kempf, a professor of mathematical physics in the Department of Applied Mathematics at ��ݮ��Ƶ. Kempf brings a wealth of experience to the role, including 10 years as Canada Research Chair in the Physics of Information. He is deeply embedded in the ��ݮ��Ƶ research ecosystem as an associate member of the Institute for Quantum Computing and the ��ݮ��Ƶ Artificial Intelligence Institute, and an associate member of the Perimeter Institute for Theoretical Physics. He also chaired the International Society for Relativistic Quantum Information for six years and currently serves on its advisory board.

Dr. Vivek Goel, president and vice-chancellor of the University of ��ݮ��Ƶ, and Dr. Reinhold Geilsdörfer, chief executive officer of the Dieter Schwarz Foundation, at the gift announcement and signing ceremony.

“I’m honoured to take on this exciting new position with support from the Dieter Schwarz Foundation,” Kempf said. “The physics of information is likely the key to understanding machine learning — and, vice versa, machine learning is likely to be the key to bringing quantum technology to maturity. That synergy holds the potential to dramatically change our way of life. It’s a privilege to lead this research and build a global team of experts in the field.”

As the Dieter Schwarz Chair, Kempf will create a global research hub grounded in strong partnerships with the Bildungscampus in Heilbronn, Germany. Mobilizing knowledge between the two regions — and others worldwide — the chair will build an elite research team at Waterloo that will include a tenure-track professor and several top graduate students and postdoctoral fellows.

Dr. Achim Kempf, the new Dieter Schwarz Chair in the Physics of Information and AI, speaks at the gift announcement.

“We’re grateful to the Dieter Schwarz Foundation for this extraordinary investment,” said Vivek Goel, president and vice-chancellor of ��ݮ��Ƶ. “The Dieter Schwarz Chair in the Physics of Information and AI is a perfect example of the ��ݮ��Ƶ at 100 vision in action. By mobilizing research and educating leaders in this field, the chair will help to ensure that our communities and partners are ready to lead in a future that promises immense opportunity to serve the needs of humanity and the planet.”

One of the largest non-profit foundations in Germany,  the Dieter Schwarz Foundation supports initiatives in science and education that prepare society to thrive in a constantly evolving future. The projects they support all connect to the Foundation’s home city of Heilbronn, which features a flourishing innovation ecosystem.

“We’re delighted to support Dr. Kempf and his research team at the University of ��ݮ��Ƶ,” said Reinhold Geilsdörfer, chief executive officer of the Dieter Schwarz Foundation. “The Dieter Schwarz Chair will facilitate an exchange of knowledge between the ��ݮ��Ƶ and Heilbronn networks that will prepare us all to embrace the future with confidence and enthusiasm. I look forward to celebrating the impact of our partnership in the years to come.”

Technological Futures

You’re better at spotting malware than you think

Ryon Jones — Tue, 05 Aug 2025 20:47:54 +0000

New study suggests that critical thinking and technical literacy are as important as cybersecurity software

When it comes to cybersecurity, humans are often seen as the weakest link, but new research suggests that with a little help, people can do a surprisingly effective job at identifying malware.

In a first-of-its-kind study, researchers from the University of ��ݮ��Ƶ’s Cheriton School of Computer Science teamed up with University of Guelph cybersecurity experts to test how users, ranging from tech novices to experts, can respond to real-time legitimate and malicious software download requests in a simulated office setting.

“Most existing malware research analyzes ‘after action’ reports, that is, investigations into what went wrong after a successful attack,” said Daniel Vogel, a professor of computer science at ��ݮ��Ƶ, and a co-author of the study. “Our study, which featured novice, intermediate and expert users, is the first malware research to observe user strategies in real time.”

Three-dozen participants received messages from fake coworkers in a Microsoft Teams-like environment, prompting them to download and install various programs. Participants had full control over whether to install the software and could research their choices however they liked.

In the initial trial, users identified malware with 75 per cent accuracy. Novice users were right 68 per cent of the time, while expert users achieved 81 per cent accuracy.

“It was interesting how novice users sometimes flagged legitimate software as malware due to a typo or poor interface design yet missed real malware when the clue was unusual system behaviour, like high processor usage,” said Brandon Lit, a PhD student in ��ݮ��Ƶ’s Cheriton School of Computer Science and the lead author of the study.

In a second round of testing, the researchers provided participants with an enhanced task manager, as well as instructions about what red flags to look for, such as software accessing large numbers of files or network connections to other countries. With that modest support, the group’s malware detection rate increased to 80 per cent.

“Just having a bit of information puts beginner users on par with computer scientists,” Lit said. “Fostering critical thinking is one of the most important things we can do to increase security.”

The study, “I’m regretting that I hit run’: In-situ Assessment of Potential Malware,” appeared in the proceedings of the 34th USENIX Security Symposium.

Math

Research

From the lab to the marketplace

Melodie Roschman — Tue, 29 Jul 2025 17:32:46 +0000

Dr. Jimmy Lin, AI researcher and Yupp chief scientist, on the vital importance of academic entrepreneurship

Is it possible for a University of ��ݮ��Ƶ professor to successfully bridge academic research and industry innovation, especially in the rapidly evolving world of artificial intelligence (AI)? The answer, it turns out, is Yupp.

Dr. Jimmy Lin is a professor of computer science and David R. Cheriton Chair in Software Systems, as well as co-director of the ��ݮ��Ƶ Data & Artificial Intelligence Institute. He is also the chief scientist at Yupp, an AI startup that launched last month with more than $33 million in seed funding, led by the veteran investor Chris Dixon at Andreessen Horowitz (a16z).

Yupp allows users to check out and compare more than 600 different AIs such as ChatGPT, Gemini, Claude, Grok, and DeepSeek for free, and in turn helps their creators improve the quality of their products. The company came out of “stealth” about a month ago with the ambition of tackling the challenges of robust and trustworthy AI evaluation. Although Yupp is still quite early, Lin and his team have already evaluated several recent AI models, including Grok-4 (from xAI) and Gemini 2.5 Flash-Lite (Google’s latest offering), revealing strong user preferences for speed.

While Yupp is based in Mountain View, California, Lin is running a ��ݮ��Ƶ office focused on research out of the Communitech hub with the help of a few intrepid co-op students.

For Lin, the relationship between industry and academia is a mutually enriching one. Lin’s goal is to align the research of his group at the University of ��ݮ��Ƶ with his work at Yupp, putting into practice this mutual exchange of knowledge to the benefit of both.

“I find great satisfaction in translating innovations from the lab into the real world and achieving broad impact,” he says. “Real-world use cases are almost always different from what you imagine in the lab. And that knowledge can be distilled to further inform your research.”

This isn’t Lin’s first foray into industry; from 2010 to 2012, he took an extended sabbatical to work at the then-fledgling Twitter (now renamed X), and since 2022 he’s also served as the chief scientist at the ��ݮ��Ƶ-based AI company Primal, which develops trustworthy enterprise-grade AI solutions.

He loves working at the University of ��ݮ��Ƶ for the quality of its students as well as the consistent focus on entrepreneurship. “If the world doesn’t care about what you’re working on, maybe it’s not that interesting to begin with,” he argues. “To me, real-world impact is the ultimate validator of research.”

Empowering partnership

Lin’s achievements, while impressive, aren’t unique. Rather, they’re the direct result of the University of ��ݮ��Ƶ’s policy of encouraging researcher entrepreneurship.

“At ��ݮ��Ƶ, we believe that empowering our researchers with ownership of their intellectual property fosters innovation, attracts top talent, and drives strong industry partnerships, ultimately translating research into real-world solutions,” says Scott Inwood, director of commercialization for the Office of Research at ��ݮ��Ƶ. “Our extensive investment in campus entrepreneurial support services further enhances this ecosystem, ensuring our researchers have the resources they need to succeed.”

That investment includes providing faculty, graduate students, and post-docs with the tools they need to build successful industry partnerships. “At ��ݮ��Ƶ, faculty aren’t just advancing research; they’re transforming it into the foundation of new companies that create real-world impact,” says John Dick, senior director of Velocity.

From bright idea to product launch, ��ݮ��Ƶ’s innovation ecosystem is there to help researchers every step of the way. “I really think that there’s something magical about ��ݮ��Ƶ: the combination of cutting-edge research, our world-renowned co-op program, and the pipelining between what happens in the lab and achieving real-world impact,” Lin says.

Math

Entrepreneurship

Watermarks offer no defense against deepfakes

Ryon Jones — Wed, 23 Jul 2025 20:34:18 +0000

��ݮ��Ƶ researchers create “UnMarker” tool that can remove any AI image watermark, highlighting continuing dangers of deepfakes

New research from the University of ��ݮ��Ƶ’s Cybersecurity and Privacy Institute demonstrates that any artificial intelligence (AI) image watermark can be removed, without the attacker needing to know the design of the watermark, or even whether an image is watermarked to begin with.

As AI-generated images and videos became more realistic, citizens and legislators are increasingly concerned about the potential impact of “deepfakes” across politics, the legal system and everyday life.

“People want a way to verify what’s real and what’s not because the damages will be huge if we can’t,” said Andre Kassis, a PhD candidate in computer science and the lead author on the research. “From political smear campaigns to non-consensual pornography, this technology could have terrible and wide-reaching consequences.”

AI companies, including OpenAI, Meta, and Google, have offered invisible encoded “watermarks” as a solution, suggesting these secret signatures can allow them to create publicly available tools that consistently and accurately distinguish between AI-generated content and real photos or videos, without revealing the nature of the watermarks.

The ��ݮ��Ƶ team, however, has created a tool, UnMarker, which successfully destroys watermarks without needing to know the specifics of how they’ve been encoded. UnMarker is the first practical and universal tool that can remove watermarking in real-world settings. What sets UnMarker apart is that it requires no knowledge of the watermarking algorithm, no access to internal parameters, and no interaction with the detector at all. It works universally, stripping both traditional and semantic watermarks without any customization.

“While watermarking schemes are typically kept secret by AI companies, they must satisfy two essential properties: they need to be invisible to human users to preserve image quality, and they must be robust, that is, resistant to manipulation of an image like cropping or reducing resolution,” said Dr. Urs Hengartner, associate professor of the David R. Cheriton School of Computer Science at the University of ��ݮ��Ƶ.

“These requirements constrain the possible designs for watermarks significantly. Our key insight is that to meet both criteria, watermarks must operate in the image’s spectral domain, meaning they subtly manipulate how pixel intensities vary across the image.”

Using a statistical attack, UnMarker looks for places in the image where the pixel frequency is unusual, and then distorts that frequency, making the image unrecognizable to the watermark-recognizing tool but undetectably different to the naked eye. In tests, the method worked more than 50 per cent of the time on different AI models – including Google’s SynthID and Meta’s Stable Signature – without existing knowledge of the images’ origins or watermarking methods.

“If we can figure this out, so can malicious actors,” Kassis said. “Watermarking is being promoted as this perfect solution, but we’ve shown that this technology is breakable. Deepfakes are still a huge threat. We live in an era where you can’t really trust what you see anymore.”

The research, “UnMarker: A Universal Attack on Defensive Image Watermarking,” appears in the proceedings of the 46^th IEEE Symposium on Security and Privacy.

Math

Research

More than $32 million awarded to advance University of ��ݮ��Ƶ research

Angelica Sanchez — Thu, 17 Jul 2025 12:00:00 +0000

The Government of Canada announces funding to support research in health, AI, climate and social equity and drive national innovation

For 17 consecutive years, the University of ��ݮ��Ƶ has been recognized as Canada's top comprehensive research university — driven by a bold, multi-disciplinary approach and strong industry partnerships that turn ideas into impact.

��ݮ��Ƶ researchers are receiving more than $32 million to support research as part of a $1.3 billion Government of Canada announcement for research across Canada.

The funding announced for ��ݮ��Ƶ researchers includes support from the Natural Sciences and Engineering Research Council of Canada (NSERC), Social Sciences and Humanities Research Council of Canada (SSHRC) and the Government of Canada’s New Frontiers in Research Fund (NFRF).

This year, more than 100 ��ݮ��Ƶ researchers from all six faculties were awarded funding through the NSERC Discovery research program, including:

Anita Layton | Faculty of Mathematics — Understanding the effects of sex and age in electrolyte homeostasis: Modelling and analysis ($685,000 over five years)
Brian Dixon | Faculty of Science — Antigen Presentation in Teleost Fishes ($480,000 over five years)
Karim Sallaudin Karim | Faculty of Engineering – Radiation detectors and systems for spectral and phase contrast X-ray imaging ($450,000 over five years)
Evan Risko | Faculty of Arts — Basic Processes in the Monitoring and Control of Cognition ($325,000 over five years)
Brian Laird | Faculty of Health – Speciation, bioaccessibility, and new approach methods for metals in wild-harvested fish in northern environments ($325,000 over five years)
Richard Petrone | Faculty of Environment — Vegetation change and water use in alpine ecosystems ($300,000 over five years)

The recent Government of Canada announcement was made by the Honourable Mélanie Joly, Minister of Industry and Minister responsible for Canada Economic Development for Quebec Regions, and the Honourable Marjorie Michel, Minister of Health supporting more than 9,700 researchers and research projects across Canada.

Numerous researchers have also received funding under the following grants and programs:

SSHRC Insight grant
SSHRC Insight Development grant
SSHRC Partnership Development grant

Several ��ݮ��Ƶ researchers received awards through the NFRF 2024 Exploration Competition, a Government of Canada program that supports high-risk, interdisciplinary projects with the potential to drive transformational innovation in areas with strong engineering and societal applications.

The recipients have demonstrated exceptional research leadership across engineering, natural sciences, health sciences and social sciences. With this federal government funding, ��ݮ��Ƶ researchers are advancing bold ideas that will transform future societies, health, economies, technologies and the environment. Learn more about their research projects below.

SSHRC Insight grant

Wendi Adair | Faculty of Arts

Understanding diversity cues and buffering social identity threat to improve representation, collaboration, and climate in Canada's diverse workplace

$258,291

��ݮ��Ƶ News - Math

Firebird forecasts and models wildfire spread

From reactive firefighting to proactive mitigation

From research idea to impact: Powered by campus supports

University of ��ݮ��Ƶ launches chair in physics of information and AI

You’re better at spotting malware than you think

From the lab to the marketplace

Empowering partnership

Watermarks offer no defense against deepfakes

More than $32 million awarded to advance University of ��ݮ��Ƶ research

SSHRC Insight grant

SSHRC Insight Development grant

SSHRC Partnership Development grant

NFRF 2024 Exploration

Canadian quantum talent helps drive global research to certify quantum communications

From biostatistics to transforming community health

The Translation Person

New Ways to Communicate

WatSPEED brings industry leaders together to advance Canada’s AI readiness

From theory to application

Building executive AI literacy

From hype to implementation: Leadership that drives change

Supporting Canada’s innovation ecosystem

New horizons for ��ݮ��Ƶ’s Canada 150 Chairs