Towards Compile-Time-Reducing Compiler Optimization Selection via Machine Learning

Tarindu Jayatilaka (University of Moratuwa)*; Johannes Doerfert (Argonne National Laboratory); Giorgis Georgakoudis (Lawrence Livermore National Laboratory); EunJung Park (Los Alamos National Laboratory); Hideto Ueno (University of Tokyo)

Branch Prediction as a Reinforcement Learning Problem: Why, How and Case Studies

Anastasios Zouzias (Huawei Technologies Switzerland AG)*; Kleovoulos Kalaitzidis (Huawei Technologies Switzerland AG, Zurich Research Center); Boris Grot (University of Edinburgh)

Custom Tailored Suite of Random Forests for Prefetcher Adaptation

Furkan Eris (Boston University)*; Marcia Sahaya Louis (Boston University); Sadullah Canakci (Boston University); Jose L. Abellan (Universidad Catolica San Antonio de Murcia ); Ajay Joshi (Boston University)

Synthesizing Low-Power Approximate Hardware with Large-Scale Search

Paras Jain (UC Berkeley)*; Safeen Huda (Google); Martin Maas (Google); Joseph Gonzalez (UC Berkeley); Ion Stoica (UC Berkeley); Azalia Mirhoseini (Google Brain)

Domain-aware Genetic Algorithm for Mapping-HW Co-optimization for DNN Accelerators

Sheng-Chun Kao (Georgia Tech)*; Michael Pellauer (Nvidia); Angshuman Parashar (Nvidia); Tushar Krishna (Georgia Institute of Technology)

ACCELNET: Learning Accelerator Design-Space

Ananda Samajdar (Georgia Institute of Technology)*; Jan Moritz Joseph (RWTH Aachen University); Tushar Krishna (Georgia Institute of Technology)

QADAM: Quantization-Aware DNN Accelerator Modeling for Pareto-Optimality

Ahmet Fatih Inci (CMU)*; Siri Virupaksha (CMU); Aman Jain (CMU); Venkata Thallam (CMU); Ruizhou Ding (); Diana Marculescu (The University of Texas at Austin)

Competition Papers

BL∪E: A Timely, IP-based Data Prefetcher

Alberto Ros (University of Murcia)*

TransforMAP: Transformer for Memory Access Prediction

Pengmiao Zhang (University of Southern California)*; Ajitesh Srivastava (University of Southern California); Rajgopal Kannan (Army Research Lab-West); Anant V Nori (Intel Corporation); Viktor K Prasanna (Unversity of Southern California)

The MPMLP: A Case for Multi-Page Multi-Layer Perceptron Prefetcher

Mehdi Saeidi (University of British Columbia)*; Ali Asgari Khoshouyeh (The University of British Columbia)*; Andrew Gunter (The University of British Columbia)*, Mieszko Lis (University of British Columbia); Prashant Nair (University of British Columbia );