 Wel, gweithio, ychydig yn fwyaf ar y sefydliad yma, ym 3 ysgol, ar Cernigamelon. Yn Keith Webster, yng Nghymru, Henry Pawsner, Yn Yr Unedig yng Nghymru, ac yw'r host yma. Yn gyfnodd, rydyn ni'n gweithio, rydyn ni'n gweithio, rydyn ni'n gweithio i chi'n gweithio ar y 9 yma, yr ystafell y 10 yma, ond mae'r pandemig ar gael yma, ac wrth gwrs, iawn ydych chi'n gweithio'r astudio cymdeithasol yn ymddiol yw'r drafodaeth yn ei ffordd a'u'r tyniol. Roeddwn i'n cael ei bod yn gweithio'r perffodus yn y Simon's Auditorium ac mae hyn yn dod i'r cyffredin iawn i'r cyffredin iawn i'r cyffredin iawn i'r cwmpatisi wedi'u gweithio'r ffordd ar y CMU Alumniau. Mae'r team yna i fath o'r ysgolio ydw i wedi cael piadau i'r CMU, a gwelio i chi'n oeswpeth hynny yna ac mae'n gweithio yw'r ffaith o'r pie ac yn fawr. Mae gweithio gweithio gweithio gyda'r llygau a phall o'n gweithio gydamerau ac mae'n gweithio'n gweithio gyda'r pie ac mae'n gweithio'n gweithio a fawr. Rhywbeth hynny yw cwmpeith gwylliantau i fynd i'w meddwl. Yn 3 munud, cyfwilio'r cwmpatyshwys yn ystod gyda'r ffaith honno'i unifiwr, y Unifiwr Gweinwysol yn Ystryliau ac yn gwneud yw'r eu glwso ddechrau arweithio yn 2008 o'r ffordd rwyf. Mae'r ffordd yn ymwybodol ddwyngol, yn ysgol Rhathiorau mewn eu cyfnod. Mae'r rheimlo ymdillodol yn cofio, mae'r cyfnodd wedi cyfnoddau mwy o'r ymdillodol yn gwneud ymddillodol ac mae'n c almonds. Mae ymdillodol yn cyfnoddau, mae'r cyfnoddau'n iawn, mewn amdillodol yn ysgolodol ac ydych chi'n gweithio'n gweithio'n gweithio'n gyngorol i'r oedden nhw i'r ddweudio'r gweithio. A gweithio'n llwythio cenedlaeth Cyniggymell yn ei ddweud o'r ddweud, ond, oherwydd y 9 yrhyw ymgylch yn ei ddweudio'r ffyniad, mae'n gweithio'n 588 ddyliadau sy'n cyfnodol ar y university. a we've had judges representing faculty and staff from colleges, schools, university libraries, members of university leadership, alumni and our wider community. 54 students participated in this year's preliminary rounds from all seven colleges, making this a truly campus-wide event. We have five of the colleges represented in tonight's finals. I'm grateful, as always, to those who are willing to help us by judging 3MT, and our finalist judges this year, sitting in the front row, are displayed on your screen at Mary Jo Dively, vice-president and general council, Anne Malloy, trustee and member of the university libraries, Dean's advocacy council, Theresa Mayer, vice-president for research, Theresa Trumbetta, assistant vice-president for alumni and constituent engagement, and Marshal Ibert, university professor and dean of the School of Computer Science. The rules are straightforward. Speakers are limited to a maximum of three minutes. Competition rules require the disqualification of anyone who continues beyond their time. Speakers will view a countdown timer, and please, if you find that the timer goes, just stop talking. We don't want to disqualify anyone. They are allowed to use a single static PowerPoint slide, but no media props, singing, dancing, anything of that sort. Our judges will consider three broad criteria in arriving at their decision. Comprehension, engagement and communication. Once all of our speakers have completed their presentations, I will escort the judges to somewhere outside so they can deliberate. During that intermission, I'll come to that in a moment. So our judging panel will select three winners receiving awards as shown on the screen, in addition to the prizes they won in the preliminary rounds. In addition, the in-person audience will have an opportunity to vote for the People's Choice Award. You should already have a ballot paper, if not, we'll make sure you get one. And we've gone for the low-tech hack-proof approach of single-paper voting. And we are also happy once again to partner with the Alumni Association to award an Alumni Choice Award. And that prize will be awarded to the participant who receives the most votes from CMU alumni watching the live stream on the Alumni Association website. The winner of the Alumni Choice Award will be announced at the end of this evening's event, along with the other award winners. Before introducing our first presenter, I'd like to thank the many people who have made this event possible. As I've mentioned, our finalists have arrived from five Heats, which together attracted 54 students. I'm grateful to all of the students from around the university who participated. Each Heat had a panel of judges, and I'm grateful to them and to the phenomenal work of my colleagues in the university libraries, especially Sonja Wellington, who has made tonight possible, and my colleagues Ryan Splenda, Sarah Young, Jimmy McKee, Shannon Riff, Sarah Bender and Shannon Baldritch. With that, I am going to move on to our first speaker, and it is my pleasure to welcome Kriti Cukkar, whose presentation is titled, Spectral Features of BCI Signals Recorded from a Stent Road. Our brain is responsible for the movements we make and the words that come out of our mouth. Now just imagine if this pathway from your brain to your body is destroyed. You will lose your ability to eat, speak, move or breathe on your own. Sounds scary, right? 30,000 people in the US alone have a disease called ALS that prevents them from making these communications. Now remember, their brain signals are still active, that is the commands are being processed, however they're not reaching the body. The important thing to remember here is that their brain signals are there, so we can record them using electrodes and convert them into something meaningful, like typing on a computer screen or asking them yes or no questions. This process of the brain interfacing directly with a computer is known as brain-computer interface. Now brain-computer interfaces have been there for a very long time. However, the ones used in ALS and the most efficient ones require open brain surgery. It is highly risky and comes with a lot of problems. Therefore, for my thesis, I am working with minimally invasive electrodes known as the stent road. Now the stent road is placed in the brain using stent technology and does not require any kind of open brain surgery. The stent road is relatively new and only six people in the US have been implanted with it till date. Once the stent road is implanted in the brain, we need to find out the spectral features of these signals, since it is new. So the participants are made to sit in front of a computer screen and asked to follow cues of movement or of rest. During the rest cue, they are asked to relax and not think about anything. But during the movement cue, they are asked to think about moving their limbs. Now remember, these participants are completely paralyzed and cannot actually move their limbs, but just the thought of it changes their brain signals. And this is what we have to find out. So we plot the power of the rest signals versus the movement signals. And in the last panel, you can see the red is the rest signal and the blue is the movement signal power. And there's a clear difference between the rest and the movement during the gamma and the high gamma frequency band. So when you ask them a question like, hey, would you like to watch a Deadpool movie? If they want to say yes, they will think about moving a limb. We will see that difference and we will know that their answer is yes. If you don't see any separation, their answer is a no. Through this groundbreaking technology, we are giving people the power to make choices on their own once again. Thank you. Our second presenter is Waku Kenoporum, whose presentation is titled Visualising the Breath of a Region, Exploring the Nexus of Air Quality, Health and Community Engagement in the Niger Delta through Art. Who remembers the wildfires and smoke from Canada last summer? Anyone? Okay, there's a few. On June 29, 2023, Pittsburgh's Air Quality Index had a rating of purple, signifying very unhealthy levels for everyone. It was our worst air pollution day in years. In fact, you couldn't see ahead of you. Now, imagine if June 29, 2023 was your reality every single day. Air pollution is a global crisis claiming seven million lives annually, with severe impacts felt most acutly in low- and middle-income countries and communities like the Niger Delta region of Nigeria that has grappled with over six decades of fossil fuel mining from companies like Shell, Chevron and Mobile. My dad, who lives in Potarkot, the larger city, laments that pollution has destroyed what's left of the Niger Delta. He says they cannot breathe because there is a hanging black cloud covering the city 24 hours of the day. But despite its devastating toll, awareness on air quality risks remain low in many communities. Traditional approaches to data dissemination often fail to engage the public effectively, especially in regions with diverse linguistic, educational and technological barriers. My research seeks to bridge this gap by employing an innovative approach that combines citizen science, technology and art to raise awareness and empower local communities to take action. Through citizen science, I'm working with local residents to deploy low-cost air quality sensors across the Niger Delta to collect real-time data on pollutants like particulate matter, ozone and carbon monoxide. Additionally, I'm gathering health data and community perspectives through interviews and surveys. But here's where the magic happens. I would translate this complex data into artistic visuals like interactive installations, paintings and sculptures. These visuals transcend language and other barriers to engage with viewers on an emotional level, making the invisible visible. And the effectiveness of these visuals will be validated through focus groups and surveys. My goal is to spark conversation, incite action and drive policy change. By empowering local communities with knowledge and agency, we can foster a safer, healthier and more sustainable future for all. Visualizing the breadth of a region represents a pioneering effort to raise air quality awareness in the Niger Delta through the intersection of science and art. Together, let's breathe more life into our communities and the planet as a whole. Thank you. Our third presenter is Arashal Rajah, whose presentation is titled Reconstructing the Neural Code for Real World Face Perception. Imagine a game of snakes and ladders with your daughter. How does your brain code for her changing facial expressions as that game unfolds? Would it surprise you that everything we know about face processing comes from lab experiments that recorded brain activity while participants looked at images of random faces on a screen? This is not how we've seen real life. We move our eyes, we choose who to look at, when to look and for how long. But we don't study the brain in the real world because capturing the environment, our behavior in it and brain activity simultaneously is incredibly difficult. In this work, we overcame those challenges by combining mobile eye tracking with brain recordings from people surgically implanted with electrodes for clinical treatment. Patients who chose to participate in this research were eye tracking glasses that recorded what they saw and where they looked at each moment over hours of natural interactions with their friends, family, clinicians and even researchers. We determined all the times they looked at faces by combining eye tracking with computer vision that also distilled all faces into a few numbers from which they could be reconstructed. Then we trained mathematical models to predict these numbers using brain activity alone and we were able to accurately predict videos of the original faces this way. Eventually, looking at the inner workings of these models revealed the brain's code for facial expressions in real life. So what is it? We found that your brain codes your daughter's expressions as deviations from her resting face. Bigger deviations mean more intense expressions. You're more sensitive to the difference between her forced smile and a relaxed smile than you are to how big the smile is. That's not all. The margin of error between intensity and type of expression is different which means your brain's code for facial expressions is oval shaped. There's even more nuance. You're more sensitive to the difference between a coi smile and a resting expression than you are to the same difference between a big smile and an even bigger smile. How do we put this scientific discovery and engineering innovation to good use? Disruptions to face processing make the simplest social communications immensely difficult for people with autism spectrum disorder. Understanding how their brain's code diverges from the baseline may help shape interventions to help them. Separately, our computational techniques can power brain-computer interfaces for vision restoration in people with blindness. There's more but just these two issues affect over 100 million people worldwide. Our fourth presenter is Benjamin Glazer, whose presentation is titled Agile Design of High Strength Aluminum Alloys. Have you ever thought about what moves us? What gets us from point A to point B in this wide world of ours? In most cases it's something like a car or a plane, right? But these systems contribute significantly to our global energy crisis being responsible for 40% of our annual CO2 emissions. This comes from the fuel and energy used to power these systems as well as to acquire, process and assemble the materials used in them. And this isn't a new problem but it's long been viewed as a mechanical issue. How do we make the engines more efficient, make them run with less fuel? But what if the problem isn't wholly mechanical? What if it's actually in the materials and that we can meet our sustainability goals by instead using lighter materials that require less fuel to use and come from more sustainable sources? Additionally, if we make them have a longer lifetime sort of need to spend more energy and fuel replacing them. But we have a candidate for this and that's aluminum. But aluminum on its own is very soft and ductile. We all know a soda can. You can pretty easily crush it, bend it or twist it and surely we don't want something like that to happen. So how do we solve this? Well we alloy the aluminum, we introduce new elements to it that resolve into strengthening phases and features. But which elements? We have dozens of ones on the periodic table and millions of possible combinations and permutations. And once we select a certain subset of elements how do we know what fractions and combinations they need to be in in order to achieve the best strengthening? It's a very high dimensional problem and our brains cannot handle it which is where my research comes in as I'm using high throughput simulations and machine learning to very rapidly explore this high dimensional data set and unravel the underlying connections between how we change the compositions of our elements and how they drive the final performance. The system I'm exploring can be strengthened in one of two ways a more conventional approach where we directly form them into our strengthening features. This causes them to develop at a micron scale the thickness of a human hair which is already very small and very strong. But for this system we can also take a more complex approach. We can first drive them into an intermediate step, activate it and drive them into our final strengthening features. This traps them at a nanometer scale a thousand times smaller bringing us even further up this strength or size curve. This system has already been used to design an alloy that's 50% stronger than a aluminum that's available on the market and as I continue to unravel the connections between the elements that make up that alloy and how we got to that final performance I can achieve ones that have up to 95% strength retention with potentially 30% reductions in cost and emissions and someday these alloys could be fueling your next set of vehicle fleets and aircraft. Thank you. Our fifth presenter is Jeremiah Milbauer whose presentation is titled Knowledge Augmentation with Artificial Intelligence. From writing to libraries to telecommunications and the web humanity has always invented new tools to organize and share knowledge. In my thesis I argue that artificial intelligence could be the next great knowledge tool if we use it to augment our intelligence rather than replacing it. So my work focuses on AI systems that help us explore the connections between ideas and across contexts. So imagine you're reading a document and every time you encounter a new idea it's highlighted and annotated. Imagine automatic fact verification embedded in every news article, explanations in every lease agreement or background info summaries when you read a scientific paper from a new field. No more helpless googling, no more endless Wikipedia tabs. Artificial intelligence has the potential to embed an expert level knowledge discovery process directly into our lives. So I began my research by looking at high level connections across communities. I used natural language processing to train conceptual maps of ideologically distinct communities and then mathematically aligned those maps to find where the communities agreed and disagreed. This technique was able to automatically discover instances of coded in-group language as well as conceptual analogies across community divides. Essentially instances where people use different terminology to talk about the same shared underlying ideas. Now these techniques could help scientists connect new ideas to their existing expertise or help all of us explore new information when we don't really know how to search for it. My work has also looked at sort of specific connections between ideas in knowledge artifacts like documents. So normally comparing documents at this very, very fine detailed level would be too computationally expensive for artificial intelligence. So I developed an algorithm which accelerates this by storing and reusing the artificial intelligence's own internal representation of each sentence. And this year as a night fellow at CMU's Center for Informed Democracy I applied those ideas to build a tool which turns news articles into documents like you see here with annotated links that connect disputed and supported information across a whole collection of related news articles. Going forward I plan to collaborate with law librarians to apply these same techniques to trace the lineage of ideas through the history of Supreme Court cases. So from law to the news to science and beyond if we use artificial intelligence to augment our intelligence we can accelerate discovery and democratize access to information for everyone. Thank you for listening. Our sixth presenter is Wengiuw whose presentation is titled The impact of opting out of standardized tests on college admissions and experimental investigation. Making college admissions decision could be life changing. According to the U.S. government statistics having a college degree earns 65% more than just a high school diploma. But now our college admissions policy is under a huge debate. Since COVID more than 80% of U.S. universities adopted the test optional policy including Carnegie Mellon meaning that students no longer need to submit their standardized test scores such as SAT and ACT. Supporters of this policy believe SAT and ACT really pose minority students at disadvantage but some schools they disagree. That mouse, Yale and MIT became the first elite universities to reinstate SAT and ACT requirement and they argue not submitting SAT and ACT scores actually hurt minority students. Despite the huge debate about test optional policy there is actually limited empirical evidence on how such policy affect individual students. So in my research I investigate this question from two perspectives. First, for students who decide to submit or withhold their SAT scores and second, from evaluator's perspective how to make decisions about students with and without SAT submitted. As a social psychology researcher I use experiments to investigate this question. In one study people play the role of high school advisors and they make decisions about whether or not to submit ACT scores for hypothetical applicants. In a subsequent study people make evaluations decisions and decide what they want to admit those applicants. What I found first, evaluators they really punish those people without SAT being submitted. Even if we tell them the score is hidden by random chance. But did applicants realize or anticipate such punishment? No, in my experiment about 40% of people withhold ACT scores and this is especially true for people lack strategic reasoning knowledge. This funding is very important. Students from low income family and under representative backgrounds they often lack guidance information about how to apply to college and they are most likely to be at disadvantage under such test optional policy. More comprehensive guidance is needed and a fair and transparent admissions policy is required in order to provide students from diverse backgrounds an equal opportunity to get access to higher education and pursue their life aspirations. Thank you. Alloys are one of the oldest and most important inventions in human history. It's formed when we combine two or more elements together to create a new material. The first alloy was made nearly 5,000 years ago. It is a combination of copper and tin. You may have heard of it, it's called bronze. It's an alloy, it's so important that an entire period of human history was named after it. Of course in the subsequent years we have made many other fundamental alloys steel, brass, cast iron, pewter, etc. While our methods of making these alloys have become more and more sophisticated we always followed one fundamental rule is that alloys have one principle element. Bronze for example is 88% copper and about 12% tin. Steel can be up to 97% iron and 3% carbon. The main element in alloy dominates and everything else is in small concentrations. This was until about the 90s when scientists asked, well what happens if we try to make a multi-principle element alloy which is an alloy with many different elements in roughly similar concentrations as opposed to having one element that dominates. Well the conventional wisdom at that time was that such an alloy would just be too unstable to exist, it would break down, it would not be of any use, there's no point thinking in this direction. But out of curiosity what scientists still persisted and not only were they successful in making such alloys I have an example here from a scientist called Fan Yang from the Northwestern Polytechnical University in China one of many examples but on testing these alloys they found that they have phenomenal physical properties high strength, high ductility great durability, great resistance to corrosion good at high temperatures and many cases these properties were better than those seen in conventional alloys like steel or bronze so this is essentially skyrocketed so it went from a bunch of scientists having hypothetical discussion to oh my god we could be replacing steel we could be replacing aluminum based alloys this is huge. So both governments and industries seeing this obviously decided to invest and as a result multi-principle element alloys today is an extremely active area of research both from the experimental standpoint where they actually synthesize and make these things in the lab and from the computer simulation standpoint which is where I work in my research I build and run state-of-the-art simulation tools that not only predict these wonderful physical properties but also actively look for more and more such alloys that could potentially change the world and this is what truly excites me so in the periodic table if you take away the toxic elements you take away the radioactive elements there are about 75 elements remaining out of which you can make alloys and this roughly translates to about 600 billion possible alloys it really begs the question what's out there that we haven't found could there be a room temperature superconductor could there be some sort of battery material that's fast charging and it has a long life you could drive from here to California and one charge or something what if there's some lightweight material you make lighter aircrafts that require less fuel I mean the possibilities are endless and these are the things that truly excite people in my field our eighth presenter is Sampada Achariah whose presentation is titled Nature-Inspired Pathogen Collection Device gotta catch em all I want everyone here to touch their closest surface and then rub their hands you might have thought that this is a very simple gesture very inconsequential and in fact you might be right over here this very gesture doesn't really have any long-lasting negative impact however if you are in an infectious environment like hospitals this very same innocuous gesture might in fact severely worsen your condition because of infections caused by pathogens present on surfaces in fact even today in the United States alone one out of every 17 patients do not survive because of this this is healthcare associated infections and clearly this is not a problem that affects a small fraction of the population this is something that can affect me you, your loved ones and every other human out there so obviously taking a sanitizer spray and spraying it on surfaces is not going to do the job we have to actively and efficiently monitor hospital surfaces a lot of work has been done in the field of pathogen detection however virtually no work has been done in the field of pathogen collection which is extremely surprising pathogen collection materials like swabs and sponges are in fact known to be unreliable inefficient something that requires a very small amount of sample area and also it is something that requires very close human interaction with the contaminated surface this is exactly why rapid and scalable surface level pathogen evaluation methodologies in order to mitigate these issues I draw my inspiration from how the nanostructures on the feet of wall climbing geckos allow them to climb on and stick to practically most surfaces on this earth based on these principles I have designed fabricated and characterised several microstructures made of soft polymeric materials as a proof of concept I was able to demonstrate that my microstructured surfaces picked up significantly higher amount of particles comparable in size to pathogens as opposed to traditional sponges and obviously because this is CMU I integrated my surfaces with a completely novel soft robot that allowed it to climb on various types of surfaces in both wet and dry conditions and even on inclines that are especially hard so now that I have these two components I plan to integrate them onto a pathogen collection device that can successfully monitor infectious environments that will help us reduce the instances of healthcare associated infections this way hospital visits can be the way they were meant to be safe and in order to make you feel better again thank you super sniffer picture a castle made of Legos sitting in front of you what if I asked you which individual pieces it's made of you would take it apart and sort each of the pieces into piles that's what chemical analysis instruments called mass spectrometers do except the Lego castle is your sample and instead of sorting Legos by shape and colour it sorts individual atoms and molecules by their mass and charge this gives us enough information to tell exactly what chemicals are in that sample mass spectrometers can only see charged particles which are called ions so the first thing they do is give a positive or negative charge to each chemical in that sample the part that does this is called the ion source however many ion sources require certain sample preparation steps followed which can be inconvenient costly in terms of time and materials and can even chemically change or destroy your sample to circumvent this we developed an ion source that allows the mass spectrometer to act as an electronic nose the super sniffer the sample simply needs to be held up to the ion source and the mass spectrometer smells it but it can smell things that we humans cannot smell for example you can take a pill hold it up to the ion source and immediately identify it as ibuprofen you can hold your finger up to it and it can smell the caffeine emanating from your skin if you've had a cup of coffee that day you can even take a dollar bill out of your pocket scan the edge of it and it can detect the trace amounts of cocaine that are present on 90% of American paper currency it works using corona discharge a highly energetic electrical plasma that emanates from the ultra sharp parabolically shaped tip of a needle that we make in the lab out of inexpensive commercially available tungsten wire when molecules emanate from the sample into the air they're sucked into the mass spectrometer and on their way they pass through the plasma where they pick up a charge and become ions we've had the pleasure of using this device for a number of projects including testing pharmaceuticals that had orbited the earth with SpaceX and non-destructively analyzing the contents of an ancient Egyptian artifact with the Carnegie Museum of Natural History down the street this simple and inexpensive device allows us to use mass spectrometry to run sensitive yet non-destructive chemical analyses directly from the air and do so on time it took me to tell you this thank you so that concludes the presentation part of this evening please join me in thanking our speakers so for those of you who are watching on the live stream and might have joined a little later reminder that if you intend to vote for the alumni choice award you can now do so through the alumni association's website for those of you here in the auditorium please vote on your ballots like this and my colleagues will be around to collect your ballots very sharply the judges and I will now retire to deliberate and we'll be back as soon as we can to announce the winners