– [Narrator] Welcome to beautiful Boulder. Located at the foot of
the Rocky Mountains, the University of Colorado Boulder has an awe-inspiring view from campus. In the Department of Computer Science innovative academic programs, hands-on opportunities,
and rigorous coursework will prepare you for the
challenges of a complex world. – Ranked 34th among engineering
colleges in the nation, the College of Engineering
and Applied Science offers the top program in
the Rocky Mountain region. At this Tier 1 research institute our graduate program in Computer Science is ranked 40th in the nation. Our faculty, staff, and students are engaged in cutting-edge
research projects that address some of the
most important challenges facing society today. Thank you for your interest in the Department of Computer Science at the University of Colorado Boulder. I hope you will join us in the future. – Founded in 1970, our department has a long history of advanced research in both basic and applied
areas of computer science, including Artificial Intelligence,
Programming Languages, Scientific Computing, and
Human Centered Computing. In addition, the University of Colorado provides an outstanding setting for interdisciplinary research. Our partnerships with campus institutes and national laboratories, such as the Institute for
Cognitive Science, BioFrontiers, and the National Center
for Atmospheric Research provide our faculty and graduate students with the opportunity to
address critical issues facing society, including
climate change, healthcare, privacy and security, and education. – There are four important factors to make your PhD studies successful and joyful; faculty, graduate program, job opportunities, and
also quality of life. And I’m proud to say that we
are strong in all four aspects. – We have excellent
faculty in our department. Our faculty have won many awards and recognitions nationally. Our PhD students have
done very well as well. We have PhD students with awards like the NSF Graduate Fellowship, the NASA Space Technology
Research Fellowship, we also have students who have won awards like the Boswell Fellowship, the Homeland Security Fellowship, and internally we have students who have won the Chancellor’s Fellowship, which is the highest of all
given to students on campus. – It’s great to be here in Boulder with so many international students. People are so welcoming
of all our cultures and they’re very supportive. – UC Boulder combines great
research opportunities with a great location. There’s a bunch of cool
restaurants here in Boulder, a growing tech industry, and
the mountains are right there. – I chose Boulder for
a variety of reasons. I am part of a great research team; I have a supportive faculty advisor; I collaborate with faculty
across the department; and I really love the outdoors here. – We offer PhD, Masters of Science, Masters of Engineering,
Bachelors of Science, and Bachelors of Arts
degree in Computer Science. Masters students may opt between a research-based option
and a course-based option. The course-based masters students may specialize in several different areas like the Algorithms,
Network, and Optimization, Data Science and Engineering, Human Centered Computing,
Intelligent Systems, Robotics, Numerical Computation, Software Systems and Cloud Computing. The course-based masters students have the option of doing a dual degree with Engineering Management Program and this degree can be done remotely. The Masters of Engineering degree is focused completely on
working professionals, and this can also be done remotely as well as a dual degree with Engineering Management Program. [mellow instrumental music] – Here, our goal is to answer language and develop useful tools for both researchers and the public. And that means we develop frameworks such as Abstract Meaning Representation to understand what a sentence means and adapt similar tools to other domains such as medical sciences
and the geosciences to help scientists make
sense of textual data sounds. In short, language is at the heart of our knowledge and our society. So, here, we harness
computational measures to understand language and to promote interdisciplinary research to realize the potential of natural
language processing. – So we want to be able to have computers understand our own spoken language, our natural language, like English. In order to do this, the computer needs to be able to understand in a sentence who is doing what to whom, and in order to accomplish
this we need humans to create these data sets
where they are marking the who and the what in these sentences. Then using machine learning, the computer can extrapolate patterns from those examples and
apply them to the new data. This aspect of having humans
to provide these examples, it costs money and it costs time. And so my area of research is looking at how we can reduce that. – We are the Theory Group, and we do math. We prove things about algorithms. In other words, we are the
backbone behind the algorithms that the computer scientists use. My specific area of research is called Spectral Graph Theory. Graphs are objects that model a lot of networks such
as Facebook and Google. They are the mathematical
abstraction of those networks. So what I do is study
graphs via the lens of math. By using Spectral Graph Theory we can infer things about the
graphs that we couldn’t do just from looking at the graph structure and by using Spectral Graph Theory we can design algorithms for graphs that we couldn’t do otherwise. – So this is the Boulder
Advanced Robotics Lab in the Computer Science Department at CU. I am the director of the Autonomous Robotics
and Perception Group here and what we focus on mostly
is experimental robotics, building platforms that actually
can autonomously navigate and do things in our environment. We have folks that work with NASA as well as some folks who have sort of started their own companies and are building different
kinds of tactile sensors. One of the projects that
I’m working on right now is associated with how
we can verify controllers on autonomous vehicles. Now what we see today are a bunch of learning-based controllers, so using machine learning
and deep learning we’re seeing a bunch of new ways in which robots can actually synthesize what they want to do in the moment. And, unfortunately, what that means for us is that there’s no way
that we can really tell based on their experiences
how they’re gonna behave. Just like a human, right? But that doesn’t really work
for safety critical systems. Autonomous vehicles, for
instance, we want them, we want to know exactly
what they’re gonna do when we start fielding them. We don’t wanna leave that up to chance. And so one of the projects
that we have going on today is National Science
Foundation funded research in how to make verified autonomy in cooperatively driving vehicles. – So at the Correll Robotics Lab, we work primarily with
swarm robotics systems in addition to humanoid robots. – We have a NASA grant on doing research on the fundamental challenges in manipulation and
perception for growing food. NASA is actually pretty much
ready to venture very far out and they are also thinking
now about being able to grow plants in space
to nurture the astronauts, making it much easier for
them to stay in space longer. And one of the big challenges
we currently tackle is to manipulate flexible objects, things like cloth, rubber tubes, or plants that will actually bend when
the robot tries to grasp them. – I work with the Droplets
Swarm Robotics Platform where currently we have
a swarm of 120 of them. We use this to test swarm
algorithms in real life, whereas most other labs
just do that in simulation. – Our systems research in the
Computer Science Department is leading the way in
developing system support for building highly adaptive applications ranging from big data analytics
to the Internet of Things. These include secure and
robust distributed systems, system support for cyber safety, and physical interfaces for
monitoring patient health. Our team is comprised of
highly innovative faculty leading the way in all aspects
of information technology. Our graduates are very well-placed in academia, industry, and research labs, and some of them have gone on to build highly innovative startups. – We have incorporated
scale of omission learning and priority scaling
techniques to help build a cyberbullying detection system that is five times more
scalable resource-wise and seven times more responsive
than the state of the art. At first it was a very
generalized centralized system, this detection system, right. But what this mobile app
will allow the parents is that they will allow
the parents to monitor their kids and on their social networks, and get notifications whenever
their kid is being bullied. – So we focus on a different area of mobile and wireless sensings, including mobile healthcare,
contact discovery, privacy and security protections, among many other mobile topics. In most recent project we work on improving healthcare practices by different wearable
devices and mobile devices that we’ve been building. We collaborated with
Colorado Children’s Hospital and School of Medicine to test our devices in a clinical setting and also partnering with industrial partners, and also forming startups to
commercialize our devices. – This is the Discovery Learning Center and this is where we focus on Human Centered Computing Research here and computer science and, in general, we focus on building technologies to support people in a
wide variety of tasks and understanding people’s needs for different types of technology. So there are a few major
labs here in this space. One is Project Epic. Project Epic focuses on
using information technology to support individuals in crises such as natural disasters
and other types of issues. The Sikuli Lab focuses
on educational technology and finding ways to support
diverse learners in learning. And the Superhuman Lab, which is our lab, focuses on building
technologies to support people with a wide range of abilities in work, in education, and in other contexts. – So I’m working on
the haptic video player which is about conveying
spatial properties in videos to people
with vision impairments. Basically, there’s a lot of
implicit spatial references and things that are not captured through any other modality
other than vision. What we’re doing is
basically putting robots directly over top of the video and allowing you to feel different motion and different translations
occurring throughout the video. If the user was blind they would basically place the robots on the
sides of the displays. Then the robots wait for an annotation, and once the annotation begins
they will actually move out and kind of speak what’s
going on in the video. The robots directly convey, in
kind of a one-to-one manner, what’s going on so that you
can feel what’s going on in the video rather than
actually look at it. – The computational science
and engineering group develops physically-based models for prediction inference and design. This is developing
mathematical techniques, numerical algorithms, often
for parallel computers that addresses problems in
science and engineering. Predicting the evolution
of fracture is seen as a grand challenge problem
in computational science. This simulation uses a
new variational technique to predict fracture. It’s allowed unprecedented
comparisons with experiments. Here in the simulation
we see a cooling process similar to what’s observed
in columnar basalt. This application is especially demanding of the algebraic solvers. The algebraic multigrade
techniques that we’ve used have enabled these simulations at much larger scale
than previously possible. – So the PLV lab, which is Programming
Languages and Verification, is mostly about making life easier for people who write code. Software today is running the, all the most important
systems in our life. Critical infrastructure,
telecommunications, medical systems, the financial market, and making sure that all
that code is safe and correct is a really big and really difficult job. The artificial pancreas
is a really cool project that our group is working on. If you have type one diabetes and you’re managing your insulin levels over the course of the day, typically what that involves today is a lot of manual checking
of your insulin level, monitoring whenever you
have a snack or a meal and that’s a real pain. So the artificial pancreas project is an attempt to develop
an automated system. The big challenge there is that you can’t really write
tests for that sort of thing because there’s so many
different possible scenarios. A group that’s working on
this project in the lab is doing mathematical
modeling to formally prove, for any possible scenario, that the machine won’t fail and that the insulin levels will stay within healthy parameters. – In the D’Mello Lab we mostly work at the intersection of cognitive science and also affective science and computing. We look at how can we model really complex behavioral processes that people do in order to
help them function better. So mostly we do that in
the context of education. So we look at how do we model
the student computationally in order to help them learn better. Right now a project that I’m working on that I’m really excited
about revolves around collaborative problem-solving. So we have students playing
a physics game together. We alternate who might be the leader in interacting with the
actual physics game, and they have to solve
some pretty hard problems. And we monitor their eye gaze, so where they’re looking at on the screen. We also monitor their
physiology to understand what their stress response might be, and we record their faces
so that we can understand what sort of facial
expressions they’re making, and we want to know how do students, in a collaborative scenario, how might their behaviors
mimic each other, and then how is that actually related to how well they do on the
task or how well they learn. – I work with the School-Wide Labs Project which is a partnership
with Denver Public Schools where we’re working on integrating physical sensing devices
into science classrooms. So we’re working with a
temperature, pressure, humidity sensor so the kids
can investigate their school. So for example we just had them go through a mold investigation where they measure the temperature and
humidity at different points in their schools looking for
the conditions for mold growth. – My work is focused on the computation that animals do on a regular basis, how they sense their local environment, process this information, and make a decision that
promotes their survival. And I’m interested in how a single animal does this computation
alone and how animals that are in the presence
of many other animals, and how this gives rise
to swarm intelligence. We are interested in how bees solve a localization problem, so how they can find the queen in a very unconstructive environment using signals that are pheromones, so they decay very
rapidly in time and space. With the swarm mechanical
response experiment what we were interested in is how a honeybee swarm is able to maintain mechanical
stability given the fact that it’s composed of many
many different individuals. So a bee can say what it’s
neighboring bees are doing but not what a bee at
the completely other side of the swarm is doing. And then what seems to be happening is that they go up the
strain of deformation so it’s actually harder
for an individual bee to maintain these mechanical stresses. But overall for the swarm that’s making things a bit easier. – My name’s Jessie. I’m a first-year PhD
student and just got elected as the Vice Chair of the
Graduate Student Association. And so we’re an
organization of technically the entire graduate student body and we have about 10-15
people that help actively plan research, social, and
inter-department communication. Every week on Fridays,
one recurring activity that we have is called Tea Time, and so we just have a
little budget for snacks and people hang out,
come play board games, just chat, take a break
from their research for an hour at a time. – The curriculum is pretty flexible here so you have a lot of opportunity
to kind choose the subjects according to your own
interest areas and all. – I like how accessible all
the faculty members are, and also how willing to
collaborate everyone is. – It’s really a great school with heavy emphasis on
interdisciplinary research and is great for just encouraging people who have not been involved in research to start working in
these different fields. – The culture in the CS department here was one where the
professors care about me, not only as a researcher but as a person. And I think my favorite
thing about being here is just the people that
I’ve been able to work with and the amount that
I’ve been able to learn. [upbeat instrumental music]

A Closer Look: CU Boulder Department of Computer Science

Leave a Reply

Your email address will not be published. Required fields are marked *