News & EventsDepartment Events & Announcements
Events
-
Sep10
EVENT DETAILS
lessOptimizing computational kernels for target GPU and accelerator hardware is an important, ongoing challenge. These kernels are the performance-critical units of AI/ML models and HPC applications, where small inefficiencies compound into large costs at scale. Traditional approaches include writing custom kernels manually, which requires significant time and expertise, as well as using compilers, which apply transformations from fixed catalogs. Recent work shows that LLM-based methods can also optimize kernels effectively, often outperforming existing compilers. PIKE, the first component of my thesis, demonstrates that in LLM-based multi-agent systems for PyTorch inference optimization, achieving a cost-effective search requires careful allocation of agents and search resources.
I propose extending PIKE to harness the strengths of tensor program superoptimization, a search-based technique that explores the space of programs within a fixed operator vocabulary to find the fastest one equivalent to the original kernel. I describe the design space that emerges from these hybrid approaches, in which a classical search technique is driven by an LLM. I further propose treating the kernel optimization search space itself as adaptable, using the LLM to modify the operator vocabulary that the superoptimizer searches over. I outline completed work on PIKE, initial steps toward integrating superoptimization as a classical search method driven by an LLM, and future directions for shaping the search in LLM-driven kernel optimization.
TIME Thursday, September 10, 2026 at 10:00 AM - 12:00 PM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Jensen Smith jensen.smith@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Sep10
EVENT DETAILS
lessHow vulnerable to input-driven attacks are the large applications used for computational science? Can these attacks be mitigated? Despite growing concerns about the security of scientific software, no work has empirically evaluated its resistance to adversarial input, and efforts to address concerns of correctness in scientific software are not currently applicable to mitigating vulnerabilities. This thesis begins to address these gaps.
I apply standard fuzzing and triage tools to demonstrate that large, open-source scientific applications are highly exploitable through their inputs with ready avenues to arbitrary code execution and thus full control over program output. Next, I implement Crucible, a novel grammar-based fuzzer that searches for small input perturbations, or “nudges”, permitting an adversary to steer the numerical results of such applications by orders of magnitude without relying on conventional vulnerabilities.
I propose an expansion of the scope of my studies to include additional applications, variants of attack, and bug detection techniques. I further propose assessing the extent to which formal specification and AI-assisted proof construction could help mitigate – or rule out – a subset of these vulnerabilities, such as the discovered numerical “nudging attacks”.
TIME Thursday, September 10, 2026 at 2:00 PM - 5:00 PM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Jensen Smith jensen.smith@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Sep14
EVENT DETAILS
lessEnjoy a welcome from Dean Christopher A. Schuh and other McCormick leaders, and receive a Northwestern Engineering T-shirt. A free breakfast on the Tech East Plaza will follow.
TIME Monday, September 14, 2026 at 9:00 AM - 10:30 AM
LOCATION LR2, Technological Institute map it
CONTACT Andi Joppie andi.joppie@northwestern.edu EMAIL
CALENDAR McCormick School of Engineering and Applied Science
-
Sep22
EVENT DETAILS
lessEnjoy a welcome from Dean Christopher A. Schuh and other McCormick leaders, and receive a Northwestern Engineering T-shirt. A free lunch on the Tech East Plaza will follow.
TIME Tuesday, September 22, 2026 at 11:00 AM - 12:30 PM
LOCATION Ryan Family Auditorium, Technological Institute map it
CONTACT Andi Joppie andi.joppie@northwestern.edu EMAIL
CALENDAR McCormick School of Engineering and Applied Science
-
Sep24
EVENT DETAILS
lesstba
TIME Thursday, September 24, 2026 at 9:00 AM - 11:00 AM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Wynante R Charles wynante.charles@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Sep29
EVENT DETAILS
lessWelcome back folks! Join us for our annual Computer Science Graduate Women's Welcome Breakfast this year on Tuesday, September 29th. Mix and mingle with graduate women across the department over a hot breakfast to kick off the year. Please RSVP to reserve your spot.
All are welcome!
TIME Tuesday, September 29, 2026 at 8:30 AM - 10:30 AM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Bella Barrios marbella.barrios@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Sep29
EVENT DETAILS
lessTuesday / CS Seminar
September 29 / 11:00 AM
Hybrid / Mudd 3514Speaker
Manuela Veloso, Carnegie Mellon UniversityTalk Title
TBAAbstract
TBABiography
TBA---
Zoom Link
Panopto LinkTIME Tuesday, September 29, 2026 at 11:00 AM - 12:00 PM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Wynante R Charles wynante.charles@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Oct5
EVENT DETAILS
lessMonday / CS Seminar
October 5 / 12:00 PM
Hybrid / Mudd 3514Speaker
Brian Suchy, Software Engineer Google DeepMindTalk Title
Google Relational Modeling Framework: Proving Relational Equivalence at ScaleAbstract
GenAI is accelerating computer science in nearly every way. From development and research to deployment, AI has the potential to enable huge sweeping changes to engineering workflows along the entire computing stack; however, to truly realize the potential for this, we must build the foundations for these workflows to be able to operate autonomously with minimal human intervention.One of these building blocks, Google Relational Modeling Framework (GRMF), generalizes relational logic, including GoogleSQL, Common Expression Language (CEL), Datalog, and (conditionally) NumPy/Pandas, into a unified Relational Algebra-based Dialect. In this talk, we dive into one of GRMF's use cases of enabling sound, deterministic verification by formally proving equivalence between relational statements.
We examine the inner workings of multi-layered formal proving methods (ranging from e-graphs and $k$-semirings to SMT solvers like Z3/CVC5) and detail real-world production use cases. Finally, we explore future applications of GRMF, including automated cross-engine query migration (e.g., BigQuery $\leftrightarrow$ F1), JIT compilation of SQL queries, execution of NumPy programs via F1/Voxel, and its potential as a general MLIR query optimizerBiography
Brian Suchy is a Senior Software Engineer at Google DeepMind. He works in several areas including: hardware development, formal verification, and (of course) AI.Brian graduated with his PhD from Northwestern in 2017. He worked in the Prescience Lab, led by Dr. Peter Dinda, with a focus on Software-based Memory Management
Research Interests: Databases, Architecture, Formal Logic
TIME Monday, October 5, 2026 at 12:00 PM - 1:00 PM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Wynante R Charles wynante.charles@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Oct14
EVENT DETAILS
lessWednesday / CS Seminar
October 14 / 12:00 PM
Hybrid / Mudd 3514Speaker
Yuhao Zhu, University of RochesterTalk Title
Imaging-Computing-Human Co-Optimizations for Visual Computing SystemsAbstract
Future computer systems will rarely operate in isolation. They will increasingly interact intimately with two domains traditionally considered outside computing: imaging and human perception. In almost all emerging applications such as Extended Reality (XR), embodied AI, and autonomous agents, computer systems must continuously sense, image, and interpret the physical world while also generating visual information that engages human perceptual and cognitive systems.This talk argues that some of the most consequential opportunities in computer systems and architecture lie at the boundaries between computing and the physical and perceptual worlds. I will illustrate this idea through two examples. First, I will discuss computational image sensors, where co-designing the sensor and computing system around domain-specific insights can substantially improve end-to-end efficiency and, more importantly, enable new forms of spatial intelligence for embodied AI. Second, I will discuss human���systems co-optimization, where we model human visual perception in immersive environments to design energy-efficient XR hardware.
Biography
Yuhao Zhu is an Associate Professor of Computer Science, Brain and Cognitive Sciences, and an Asaro Biggar Family Fellow in Data Science, all at University of Rochester, where he and his students are mostly drawn to research problems at the intersection of computer architecture, graphics, and human factors. With his collaborators and students, he has won multiple IEEE Micro Top Picks in Computer Architecture for ���novelty and potential for long-term impact���, along with best paper awards/nominations at MICRO, HPCA, ASPLOS, and IEEE VR.Research Areas: Computer architecture, graphics, imaging
TIME Wednesday, October 14, 2026 at 12:00 PM - 1:00 PM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Wynante R Charles wynante.charles@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)
-
Oct29
EVENT DETAILS
lesstba
TIME Thursday, October 29, 2026 at 9:00 AM - 11:00 AM
LOCATION 3514, Mudd Hall ( formerly Seeley G. Mudd Library) map it
CONTACT Wynante R Charles wynante.charles@northwestern.edu EMAIL
CALENDAR Department of Computer Science (CS)