Date | Discussion Lead | Paper/Topic | Recording |
---|---|---|---|
Jan 29 | Charith Mendis | Organization Meeting | |
Feb 5 | Marie McCord | Search-Based Regular Expression Inference on a GPU | Video |
Feb 12 | Neil Zhao | Modern Clouds: Side-Channel Attacks and Defenses | Video |
Feb 19 | Jacob Laurel | Automated Analyses for Continuous Computations | Video |
Feb 26 | Joshua Levine | Efficiently Scaling Transformer Inference | Video |
Ahan Gupta | A Framework for Fine-Grained Synchronization of Dependent GPU Kernels | ||
Mar 4 | Yilin Zhang | Efficient Parallel Functional Programming with Effects | Video |
Mar 11 | Spring Break (no class) | ||
Mar 18 | Paul Krogmeier | Learning Symbolic Concepts and Domain-specific Languages | Video |
Mar 25 | Charles Yuan | Building the Tools to Program a Quantum Computer | Video |
Apr 1 | Muyan Hu | Optimal Kernel Orchestration for Tensor Programs with Korch | Video |
Apr 8 | Vimarsh Sathia | Exocompilation for productive programming of hardware accelerators | |
Kun Wu | Hector: An Efficient Programming and Compilation Framework for Implementing Relational Graph Neural Networks in GPU Architectures | ||
Apr 15 | Isha Chaudhary | COMET: Neural Cost Model Explanation Framework | Video |
Jiawei Liu | NeuRI: Diversifying DNN Generation via Inductive Rule Inference | ||
Apr 22 | Yufeng Wang | TGLite: A Lightweight Programming Framework for Continuous-Time Temporal Graph Neural Networks | Video |
Yifan Zhao | Felix: Optimizing Tensor Programs with Gradient Descent | ||
Apr 29 | Steven Xia | Fuzz4All: Universal Fuzzing with Large Language Models | Video |
Zhihao Wang | Efficient GPU Kernels for N:M-Sparse Weights in Deep Learning | ||
Rahul Bothra | Formal Abstractions for Packet Scheduling | ||
May 6 | Finals Week (no class) |
Date | Discussion Lead | Paper/Topic | Recording |
---|---|---|---|
Aug 21 | Charith Mendis | Organization Meeting | |
Sept 4 | Labor Day (no class) | ||
Sept 11* | Yifan Zhao | ApproxCaliper: A Programmable Framework for Application-Aware Neural Network Optimization | Video |
Sept 18 | Stefanos Baziotis | A Fast In-Place Interpreter for WebAssembly | Video |
Haoyang Zhang | RipTide: A Programmable, Energy-Minimal Dataflow Compiler and Architecture | ||
Sept 25* | Randy Huang, Hongbin Zheng (AWS) | A Deep Learning Compiler for Systolic Array based Accelerators | |
Oct 2 | Jacob Laurel | Abstractly Interpreting Differentiable Programming | Video |
Oct 9 | Jason Ansel (Meta) | PyTorch 2.0: Faster Machine learning through dynamic Python bytecode translation and graph compilation | Video |
Oct 16 | Adrian Sampson (Cornell) | The Next 700 Accelerator Design Languages | Video |
Oct 23 | Rohan Yadav (Stanford) | Distributed Sparse Computing in Python | |
Oct 30 | Muchen Xu | Hydride: A Retargetable and Extensible Synthesis-based Compiler for Modern Hardware Architectures | |
Nov 6 | Vir Narula | Pin or Fuse? Exploiting Scratchpad Memory to Reduce Off-Chip Data Transfer in DNN Accelerators | |
Muyan Hu | EinNet: Optimizing Tensor Programs with Derivation-Based Transformations | ||
Nov 20 | Fall Break (no class) | ||
Nov 27 | Jaeyeon Won (MIT) | The Continuous Tensor Abstraction: Where Indices are Real | |
Dec 4 | Milind Kulkarni (Purdue) | Beyond Polyhedra: Optimizing Irregular Programs | |
Dec 11 | Finals Week (no class) |
*5–6 pm in Room 2405 (Siebel)
*Thursday March 2nd, 12:30–1:30 pm in Room 1302 (SC)
Date | Discussion Lead | Paper/Topic | Recording |
---|---|---|---|
Oct 11 | Organization Meeting | ||
Oct 18 | Charith, Vikram, Sasa, Gagandeep, Lawrence | Panel Discussion | |
Oct 25 | Jacob Scott Laurel | Statheros: A Compiler for Efficient Low-Precision Probabilistic Programming | |
Nov 1 | Gagandeep Singh | Learning Fast and Precise Numerical Analysis | |
Nov 8 | Aayan Kumar | Shiftry: RNN Inference in 2KB of RAM | |
Nov 15 | Kun Wu | PyLog: An Algorithm-Centric Python-Based FPGA Programming and Synthesis Flow | |
Nov 29 | Hanchen Ye | Automatic generation of high-performance quantized machine learning kernels | Video |
Dec 6 | Adel Ejjeh | HPVM2FPGA: Enabling True Hardware-Agnostic FPGA Programming |