CSE
CSE
CSE CSE

2018-2019 SURE Research Projects in Computer Science and Engineering (CSE)

This page lists summer research opportunities in CSE that are available through the SURE Program. To learn more or apply, visit: https://sure.engin.umich.edu/.

Directions:

  • Please carefully consider each of the following projects, listed below, before applying to the SURE Program.
  • You must indicate your top three project choices on your SURE application, in order of preference, using the associated CSE project number.
  • Questions regarding specific projects can be directed to the listed faculty mentor.

CSE Project #1: A Secure Architecture Based on Ensembles of Moving Target Defenses (Computer Security Project)
Faculty Mentor: Todd Austin [austin @ umich.edu]
Prerequisites: EECS 280, C++ programming, compiler/operating system experience.
Description: Secure systems today are built by identifying potential vulnerabilities and then adding protections to the system to thwart the associated attacks. Unfortunately, the complexity of today's systems makes it impossible to prove that all attacks (or even a class of attacks) can be stopped, so clever attackers find a way around even the most carefully designed protections. In this work, we take secure system design on the offensive - instead of attempting (and failing) to find every last vulnerability, we create an impenetrable sequence of roadblocks between the attacker and the critical information assets they require to mount an attack, a technique we call Ensembles of Moving Target Defenses (EMTDs). The project is building a complete system based on EMTD protections, which will involve building architecture, microarchitecture, compiler, operating system, and runtime support.

CSE Project #2: Into the AI Void (Computer Architecture Project)
Faculty Mentor: Todd Austin [austin @ umich.edu]
Prerequisites: EECS 280, EECS 370, C++ programming, compiler/operating system experience.
Description: As everyone (and their brothers, and sisters, and aunts, and cousins) works on deep neural nets, machine learning, and AI algorithms, it creates a void surrounding research on applications that do NOT lend themselves to well to neural nets and machine learning. In this project, we will be performing a deep-dive on the capabilities of neural nets and machine learning to understand what they CANNOT do well, and build understanding as to what specialized hardware support can be created for potential post-AI algorithms. We've already identified a number of promising candidates, and we are eager to get more hunters on the trail!

CSE Project #3: Smart Agents that Know their Users
Faculty Mentor: Nikola Banovic [nbanovic @ umich.edu]
Prerequisites: Strong programming background; experience with Machine Learning and Reinforcement Learning is a plus.
Description: The goal of smart agents that run on personal and mobile devices (e.g., Cortana, Google Assistant, Siri) is to provide people with relevant and personalized information and assist them in everyday tasks. Such agent automatically learning about the users over time by observing them and inferring information about them (e.g., using methods such as Active Learning). However, sensing and learning user information takes time. This leads to significant downtime between when the user starts using the agent and when the agent has learned enough about the user to offer personalized services. Furthermore, existing methods make unrealistic assumptions that the users will answer unlimited number of questions without refusing to answer any of those questions and always answering truthfully. In this project, we design, implement, and study smart agents that can learn which questions to ask about the users and when to ask them to learn as much as possible about the users. The goal of our agents is to maximize their adoptability and minimize the users' attrition and the burden and inconvenience of asking socially unacceptable or awkward questions. Knowledge from this project will inform the design of future smart agents that can automatically learn fundamental user properties to provide them with a starting point for inference about the specific information they need to make their specialized predictions and recommendations.

CSE Project #4: Boosting the performance of graph-based algorithms
Faculty Mentor: Valeria Bertacco [valeria @ umich.edu]
Prerequisites: EECS 281, EECS 370. Recommended: C++, scripting, EECS376.
Description: More and more applications rely on graphs as the underlying data structure: from social networks, to internet's web connections, to brain's neurons and geo maps, and even consumers' product preferences. The performance of these algorithms is often limited by the latency of accessing vertices in memory, whose access present poor spatial locality. The goal of this project is to boost the performance of graph-based algorithms by developing hardware and software solutions to this end: we plan to work on the data layout, on ad-hoc data structures and on designing dedicated hardware acceleration blocks. We hope to boost the performance of graph traversals and computation by 3-5x.

CSE Project #5: Counter-attacking security attacks
Faculty Mentor: Valeria Bertacco [valeria @ umich.edu]
Prerequisites: EECS 280, EECS370. Recommended: C/C++, scripting.
Description: The biggest challenge in the Internet-of-Thing (IoT) industry is security. Because of their low cost, IoT device designers put little effort in providing and designing security features in their products -- making them a very easy target for security attackers. In addition, other products with widespread deployment, have just recently become "connected" and security has not been at the forefront of concerns. Popular and recent attacks include baby cams, automobiles, pacemakers, and airline jets. Instead of developing protections against security attacks, this project proposes to design a counter-attack: the more an attacker probes our system to break in, the more we change its software and hardware stack to throw the attacker off.

CSE Project #6: Find out how fast is your new shiny hardware device -- without building it
Faculty Mentor: Valeria Bertacco [valeria @ umich.edu]
Prerequisites: EECS 280, EECS370 or EECS 373. Recommended: C/C++, scripting.
Description: Technology is moving towards the design of complex hardware systems, comprising many engines and accelerators: a common example is self-driving cars, which require lots image, video and sensor data processing, vehicle-to-vehicle communication, actuators control, etc. Traditionally, the performance of these systems could only be evaluated by building expensive prototypes -- but then mistakes are very costly. This project is centered on building a framework that allows simulation of complex heterogeneous hardware systems comprising many hardware units and accelerators, so as to evaluate its performance and optimize the design of the software, interconnect, and even the selection of hardware components in the systems.

CSE Project #7: Extracting "computational patterns" from apps to get their best performance out
Faculty Mentor: Valeria Bertacco [valeria @ umich.edu]
Prerequisites: EECS 281. Recommended: C++, python.
Description: Performance is the key metric of success for data-intensive software applications. As the benefits of transistor scaling are coming to an end, heterogeneous computing systems hold the promise to keep up the continuous performance improvements we all have become used to. Heterogeneous systems comprise many different hardware accelerators (CPUs, GPUs, FPGAs, machine learning, and crypto engines etc.), each specialized to accelerate one computation kernel. But how can we extract all the relevant kernels from a software applications, so that we can run each part on its best fitting hardware unit? The goal of this project is to analyze a software application's source code to extract portions of code / patterns of computation that are best suited for mapping to each hardware accelerator.

CSE Project #8: Boosting the performance of graph-based algorithms
Faculty Mentor: Valeria Bertacco [valeria @ umich.edu]
Prerequisites: EECS 281, EECS 370. Recommended: C++, scripting, EECS 376.
Description: More and more applications rely on graphs as the underlying data structure: from social networks, to internet web connections, to brain neurons and geo maps, and even consumers' product preferences. The performance of these algorithms is often limited by the latency of accessing vertices in memory, whose access present poor spatial locality. The goal of this project is to boost the performance of graph-based algorithms by developing hardware and software solutions to this end: we plan to work on the data layout, on ad-hoc data structures and on designing dedicated hardware acceleration blocks. We hope to boost the performance of graph traversals and computation by 3-5x.

CSE Project #9: Millimeter-scale Sensor System
Faculty Mentor: David Blaauw [blaauw @ umich.edu]
Prerequisites: Interest in embedded systems, circuit design. Responsibilities will depends on candidate background.
Description: In this project we are looking for a student to help us with embedded hardware and software development of mm-scale sensor systems for Internet of Things (IoT) applications. In the last few years we have prototyped the world's first complete and functional mm-sized embedded systems. The system incorporated a commercial ARM Cortex M0 processor with low leakage memory, ultra-low power flash, battery, harvesting and RF communication. Early versions sensed temperature and pressure and more recent versions can also record audio and images. This work is currently featured at the Computer History Museum in California as the world's "smallest computer" and also in the atrium of the EECS building. Our team is currently working actively with other researchers to deploy the sensors for butterfly tracking, "down-hole" oil-reservoir exploration and medical implantable applications as well as commercializing the technology. We are currently working to expand our sensor capabilities to include new sensing modalities, better interfaces and increasing radio range. The student work will depend on the background of the candidate and can include embedded software development for low power operation, GUI design, development of new sensor applications, help with digital and mixed signal circuit design, and testing and diagnosis of fabricated chips.

CSE Project #10: Accelerating Whole Genome Sequencing
Faculty Mentor: David Blaauw [blaauw @ umich.edu]
Prerequisites: Interest in biomedical algorithm design, FPGA implementation, VLSI Verilog chips design. Responsibilities will depends on candidate background.
Description: Whole genome sequencing (WGS) determines the complete DNA sequence of an organism's genome. While it cost nearly $3 billion to sequence the first human genome in 2001, just over the last one decade, the production cost of sequencing has plummeted from ten million dollars to thousand dollars, making it a promising tool for individualized treatment plans and precision health. In the sequencing pipeline, hundreds to thousands of CPU hours of intensive computation needs to be performed on raw data to sequence one genome, which are opportunities for software algorithm optimization and hardware acceleration. In this project, we are developing novel algorithms tailored for hardware acceleration to speed up the entire secondary analysis for a number of emerging genomics applications, such as whole genome sequencing, RNA single cell sequencing, and microbiome analysis. We aim to build a heterogeneous hardware system to speed up the current software pipeline by 1000x or more using algorithm and FPGA/ASIC implementation co-design. The work will depend on the background of the candidate and can involve both software / hardware development as well as exploring new genomics applications.

CSE Project #11: Crowd-Sourced Vision-Based Pollution Sensing
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: Prior experience with machine learning techniques.
Description: Develop methods of estimating air quality based on crowd-sourced data, potentially gathered for unrelated reasons. We are currently developing techniques allowing outdoor photographs posted on social media sites for any reason to be analyzed in order to estimate concentrations of different types of air pollution, including particulates and NO2. This work requires developing understanding of pollutant optical properties, computer graphics, machine vision, and machine learning. It also provides a smooth path from (relatively straight-forward) data gathering and analysis to the development of original ideas.

CSE Project #12: General-Purpose IoT Gateway
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: PCB design and testing.
Description: Develop a general-purpose Internet-of-Things gateway that can easily and practically be used by scientists to turn manual sensing systems into automatic, very long-range wireless sensing systems. We are interested in this problem because we have frequently encountered researchers in several fields including Civil Engineering, Air Quality Research, and Entomology looking for exactly this solution. We currently have a prototype design, and are in the process of iteratively testing and assembling components. An undergraduate researcher would initially help with prototyping and testing the current design, then advance to improving it, completing the printed circuit board layout, and assisting on developing firmware.

CSE Project #13: Crowd-sourced Photo-based Air Quality Estimation
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: Programming for data analysis. Algorithm design. Basic understanding of optical scattering and absorption.
Description: Develop methods of estimating air quality based on crowd-sourced data, potentially gathered for unrelated reasons. We are currently developing techniques allowing outdoor photographs posted on social media sites for any reason to be analyzed in order to estimate concentrations of different types of air pollution, including particulates and NO2. This work requires an understanding of pollutant optical properties, computer graphics, machine vision, and machine learning. It also provides a smooth path from (relatively straight-forward) data gathering and analysis to the development of original ideas.

CSE Project #14: Sensing and Understanding Natural Environments
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: Comfort with using mathematical algorithms and data analysis. Python knowledge a plus.
Description: Help design a wireless ultra long battery life system for automatically sensing and understanding natural audio environments. The focus is on classifying and counting pollinators. Help figure out why so many bee colonies have died that there has been a 20% increase in the cost of pollination services, which accounts for billions of dollars per year, and why there has been a 75% reduction in the mass of flying insects in nature reserves in the base 25 years. Initially, participants will collect and analyze audio data on bee activity. The goal is to identify the presence and classify bee genus and activity to support entomological research. Ultimately, low-power IoT devices will be designed for deployment in agricultural environments.

CSE Project #15: Infrastructureless Communication Smartphone App
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: Smartphone programming or marketing experience.
Description: How would you like to be put back in control of what you see and share via social media, even as others attempt to use it as a tool of censorship and surveillance? Develop Android/iOS microblogging applications supporting direct and transitive phone-to-phone communication that is resistant to outages, blocking, censorship, and surveillance. Create local community-oriented networks that would keep working even if the plug were pulled on the internet. Development team members will need iOS or Android programming experience. Marketing team members will need website design, video production, or social media marketing experience.

CSE Project #16: Smart IoT Sensor Interface Design
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: One or more of the following. 1) PCB design. 2) PCB assembly, including surface mount soldering. 3) Electrical system testing and characterization. 4) Analog circuit design.
Description: Implement and test a smart sensor interface that turns dumb, manual sensors into accurate, automatic, wireless remote sensing systems. Water quality, air quality, audio... almost anything. Make the IoT real, today. Academic users/customers already enthusiastically waiting for the product.

CSE Project #17: IoT Communication Design and Modeling
Faculty Mentor: Robert Dick [dickrp @ umich.edu]
Prerequisites: 1) High-level understanding of wireless transceiver power consumption characteristics and ability to learn. 2) Ability to measure circuit power consumption. 3) Ability to interface various transceivers with development boards, e.g., Arduinos, using serial communication protocols.
Description: The personal computer, internet, and mobile computing changed mankind. The Internet of Things (IoT) is next. This network of ubiquitous, often low-power and wireless, devices will sense, analyze, and control the world. To create it, system designers must see the implications of their design decisions, and to do that they need system-level models of the wireless communication transceivers they are considering. There are many competitors (LoRa, NB-IoT, Weightless W, N, and P, Sigfox, and others), most of which are new, and nobody knows which will succeed.

CSE Project #18: Computer Vision for Physical and Functional Understanding
Faculty Mentor: David Fouhey [fouhey @ umich.edu]
Prerequisites: Good grades in EECS 442 OR EECS 445
Description: The lab is broadly focused on building 3D representations of the world and understanding human/object interaction. Potential projects include learning about: navigating environments, object articulations, commonsense physical properties of objects, and hand grasps. Please look at: http://web.eecs.umich.edu/~fouhey/ for a sense of what projects we've done in the past. We will find a specific project based on mutual interest and particular abilities (e.g., stronger systems programming abilities, experience with graphics, etc.). Students looking for a longer term project continuing during the school year are strongly encouraged to apply.

CSE Project #19: Software development for applications of discrete event systems to control, security, and privacy
Faculty Mentor: Stephane Lafortune [stephane @ umich.edu]
Prerequisites: Programming experience in C, C++ and/or Java required.
Description: Discrete Event Systems (DES) are a class of dynamical systems that occurs widely in modern technology, such as in Cyber- and Cyber-Physical Systems. The UMDES Group does research on several aspects of DES modeled by finite-state automata, including supervisory control, fault diagnosis, detection of sensor deception attacks, privacy enforcement, and so forth. Our group has been working several software tools that implement the algorithms developed in our research.

UMDES is a library of routines, written in C or C++, that implement the most common algorithms for supervisory control, diagnosis, and opacity analysis. DESUMA is Graphical User Interface that embeds UMDES commands and draws the layout of small to medium-sized automata. See: wiki.eecs.umich.edu/desuma for further details about these tools.

Other tools under current development are listed at:
https://gitlab.eecs.umich.edu/M-DES-tools
and include VEiP:
https://gitlab.eecs.umich.edu/M-DES-tools/VEiP

The SURE intern will work on improving and adding new functionalities to the above tools. This will include the development of case studies. The intern will work closely in collaboration with the members of the UMDES Group.

CSE Project #20: Human-AI Teams
Faculty Mentor: Walter Lasecki [wlasecki @ umich.edu]
Prerequisites: Have some programming experience, be willing to learn about study methods and web programming, the ability to work on tasks both independently and with groups.
Description: The most powerful and generalizable systems are not AI or human, but a combination of the two. To make human-AI teams effective, it is important to be able to understand what the system knows / is capable of, and anticipate where to trust it. We leverage knowledge of human interaction in psychology, cognitive science, organizational behavior, management science, and other fields to better understand how AI agents can interactively support and augment people's tasks. In both one-on-one and team settings, we are exploring how AI can be not only a source of distributed cognition (a support tool with memory) but also of team cognition, such that we can build human-AI team mental models for seamless interaction about tasks and goals. We aim to find effective ways of building tools that help people form these shared mental models and transactive memory systems. We are studying how human and AI agents working on shared tasks form an understanding of the others' abilities, and how this can affect joint performance — especially as these models continue to change as systems learn.

CSE Project #21: AR Collaboration
Faculty Mentor: Walter Lasecki [wlasecki @ umich.edu]
Prerequisites: Have some programming experience, be willing to learn about study methods and web programming, the ability to work on tasks both independently and with groups.
Description: How can augmented and virtual reality be used to create more powerful, effective collaborative experiences? This project aims to introduce techniques and systems for designing new collaborative tools and environment on-the-fly as people collaborate. If successful, we will be able to shed the traditional limitations of physical reality when it comes to on-demand creation of useful artifacts, and improve people's ability to collaborate both in shared spaces and remote (virtual) spaces.

CSE Project #22: Deep learning for perception and reasoning
Faculty Mentor: Honglak Lee [honglak @ umich.edu]
Prerequisites: Grade of A or higher in undergrad ML (EECS 445); solid skills in calculus, probability, and statistics; strong programming skills; prior experience in deep learning libraries (Tensorflow, pytorch, etc.) is desirable but not required.
Description: The brain has an impressive ability to process a variety of sensory input data, including images, sounds, languages, and touches. Recent biological and computational studies suggest that the brain may be using a single machine learning algorithm to develop representations from such diverse sensory domains. Furthermore, humans can readily learn from vast amounts of unlabeled data, together with only a small amount of supervision; this is because humans can easily recognize and discover underlying structures from seemingly complex input data. Inspired by this evidence, we aim to develop machine learning algorithms to develop good feature representations from large unlabeled and labeled data. In this project, we will focus on developing machine learning algorithms and applying them to perception and reasoning problems, which involve computer vision and language. Students who intend to continue beyond summer and perform long-term research (at least a year) are strongly encouraged to apply.

CSE Project #23: Deep reinforcement learning
Faculty Mentor: Honglak Lee [honglak @ umich.edu]
Prerequisites: Grade of A or higher in undergrad ML (EECS 445); solid skills in calculus, probability, and statistics; strong programming skills; prior experience in deep learning libraries (Tensorflow, pytorch, etc.) is desirable but not required.
Description: The brain has an impressive ability to process a variety of sensory input data, including images, sounds, languages, and touches. Recent biological and computational studies suggest that the brain may be using a single machine learning algorithm to develop representations from such diverse sensory domains. Furthermore, humans can readily learn from vast amounts of unlabeled data, together with only a small amount of supervision; this is because humans can easily recognize and discover underlying structures from seemingly complex input data. Inspired by this evidence, we aim to develop machine learning algorithms to develop good feature representations from large unlabeled and labeled data. In this project, we will focus on developing machine learning algorithms and applying them to perception and control problems, for example where an agent interacts with a complex environment with partial observability and sparse rewards. In order to tackle the problem better, we will also explore state-of-the-art techniques in vision and language. Students who intend to continue beyond summer and perform long-term research (at least a year) are strongly encouraged to apply.

CSE Project #24: Cloud-Driven 3D Printing
Faculty Mentor: Harsha V. Madhyastha [harshavm @ umich.edu]
Prerequisites: EECS 482
Description: Today, the execution of a 3D printer is driven by a locally running controller which translates a high-level geometric specification of the product being printed into low-level commands sent to the device's motors. This design suffers from two fundamental limitations: the local controller's execution is constrained by the amount of compute power on the device, and the controller is hard to update when new algorithms for it are developed.

To address these limitations, in this project, we will realize a new architecture in which the execution of a 3D operation will be managed by a controller running in the cloud. This offers practically infinite compute resources to the controller, makes it straightforward to update the controller's code, and also enables collaborative optimization (e.g., the controller can learn from the operation of some devices to fine-tune the operation of other devices of the same type).

Some of the challenges that will need to be tackled in realizing this new architecture include: a) making the cloud controller fault-tolerant (i.e., switching from a cloud controller in one region to another based on current Internet path characteristics), and b) enabling the local controller to take over safely when the device loses Internet connectivity.

CSE Project #25: Quantifying Broken References on the Web
Faculty Mentor: Harsha V. Madhyastha [harshavm @ umich.edu]
Prerequisites: Required: EECS 281 + Familiarity with a scripting language, such as Python; Desirable: EECS 485
Description: Today, much of human knowledge is documented publicly on the web in terms of blog posts, discussions on forums, news articles, etc. Therefore, decades or even centuries from now, one can envision archaeologists learning about the human past by crawling the web, instead of having to explore the world to find hidden scriptures and inscriptions.

However, web crawls conducted at a future point in time are likely to find that many of the links on the web lead to broken references. The web is intrinsically an inter-linked structure with pages often linking to other pages for background, context, or reference. As domains expire and as content is purged, many of these links are rendered broken.

In this project, our goals will be to:
1) quantify the extent to which the web is broken today in terms of references that lead to nowhere
2) characterize the reasons for these broken references and their impact on one's ability to understand the content on a page
3) design and develop potential solutions

CSE Project #26: Computational Medicine: A Gateway to Computer-Aided Decision-Making
Faculty Mentor: Kayvan Najarian [kayvan @ umich.edu]
Prerequisites: Linear Algebra, Machine Learning, Programming Experience
Description: The primary research focus of the Biomedical and Clinical Informatics Lab (BCIL) is the development of computer-assisted decision-making systems. Using insightful biomarkers and underlying patterns extracted from raw data via signal processing, image processing and machine learning, these systems can provide real-time, on-the-fly treatment recommendations and outcome predictions at every stage of care.

The undergraduate researcher will learn advanced machine learning techniques and apply them to a variety of clinical decision support systems currently being developed. During their SURE tenure, the researcher will have the opportunity to work with and learn from a diverse group of mathematicians, computer scientists, and clinicians who comprise the BCIL team. They will also have the opportunity to continue their research work within the lab after the summer term.

CSE Project #27: The Internet of Everything: Bringing "dumb" object to the digital world with RFID tags
Faculty Mentor: Alanson Sample [apsample @ umich.edu]
Prerequisites: Good programming skills
Description: RFID tags are battery-free, paper-thin stickers that can communicate with RFID readers form +8 meters of distance. These tags offer a minimalistic means of instrumenting everyday objects. By monitoring changes in the low-level communication channel parameters between the tag and reader, it is possible to turn an RFID tag into an ultra-low cost, battery-free sensor. Applications include in-home activity inferencing, interactive physical objects, and human-robot interaction.

CSE Project #28: Algebraic Problems and Algorithms
Faculty Mentor: Ilya Volkovich [ilyavol @ umich.edu]
Prerequisites: EECS 376, knowledge of abstract algebra – ideally familiarity with finite fields, proficiency in proof-based mathematics.
Description: This is an opportunity for undergraduate students to get exposure to research conducted in the intersection area between mathematics and computer science. More specifically, the roles of randomness and algebra in theory of computations and algorithms.

In recent years, many combinatorial and graph-theoretical algorithms very developed leveraging the nice algebraic structure of the underlying problems. Yet, there is still many problems that could see an improvement using more sophisticated analysis. The goal of this project is to attempt to improve the running time of some previously existing algebraic algorithms.

CSE Project #29: Computational Strategic Reasoning
Faculty Mentor: Michael Wellman [wellman @ umich.edu]
Prerequisites: Programming ability; interest/background in finance, economics, game theory, and/or statistics (helpful though not required)
Description: The Strategic Reasoning Group (strategicreasoning.org) develops computational tools to support reasoning about complex strategic environments. Recent applications include scenarios arising in finance and cyber-security. We employ techniques from agent-based modeling, game theory, and machine learning.

CSE Project #30: Deep Learning Techniques for Forecasting Trajectories of Ocular Disease
Faculty Mentor: Jenna Wiens [wiensj @ umich.edu]
Prerequisites: EECS 445
Description: Our team is working to extract detailed data of structures in the back of the human eye (retina, optic nerve, blood vessels) that is routinely captured in photographs and other ocular imaging modalities. We are looking to integrate this data, along with clinical data from visits to eye care professionals captured in electronic health records into deep learning algorithms that can be used to forecast the trajectory and outcomes of patients with ocular diseases such as glaucoma, macular degeneration, and diabetic retinopathy. In this project, you will work closely with the faculty mentor and a clinical collaborator: Dr. Joshua Stein.

CSE Project #31: Machine Learning for Data-Driven Decisions
Faculty Mentor: Jenna Wiens [wiensj @ umich.edu]
Prerequisites: EECS 445 (machine learning), linear algebra, probability and statistics, strong programming skills
Description: We focus on problems that lie at the intersection of machine learning and healthcare. We have several projects that leverage time-series data, including structured clinical data, unstructured clinical notes, physiological waveforms, and images, for building predictive models of adverse outcomes and matching patients to treatments. Our applications range from short-term in-hospital outcomes (e.g., acquisition of healthcare-associated infections) to long-term chronic diseases (e.g., progression of cystic fibrosis). While we are motivated by problems with potential for impact in healthcare, we make contributions to machine learning through the development of novel techniques for organizing, processing and transforming these data into actionable knowledge. To this end, we are particularly interested in time-series analysis, transfer/multitask learning, intelligible models, and novel techniques for incorporating domain expertise. In this project, the undergraduate researcher will support the development and analysis of computational tools for improving healthcare. Students who intend to continue research beyond the summer are especially encouraged to apply.