More Like this

1. Improving the Speed of gem5’s GPU Regression Tests

   James Braun; Matthew D. Sinclair (June 2023, 5th gem5 Users' Workshop)

   In recent years, we have been enhancing and updating gem5’s GPU support. First, we have enhanced gem5’s GPU support for ML workloads such that gem5 can now run. Moreover, as part of this support, we created, validated, and released a Docker image that contains the proper software and libraries needed to run GCN3 and Vega GPU models in gem5. With this container, users can run the gem5 GPU model, as well as build the ROCm applications that they want to run in the GPU model, out of the box without needing to properly install the appropriate ROCm software and libraries. Additionally, we have updated gem5 to make it easier to reproduce results, including releasing support for a number of GPU workloads in gem5-resources and enabling continuous integration testing on future GPU commits. However, in an effort to provide sufficient coverage, the cur- rent testing support for GPU tests requires significant runtime both for the nightly and weekly regression tests. Currently most of these regression tests test the GPU SE mode support, since GPU FS mode support is still nascent. Unfortunately, much of this time is spent parsing input files to create arrays and other data structures that the GPU subsequently computes on. Although SE mode does not simulate the system calls needed to read these input files, nevertheless this still represents a significant overhead that increases runtime and prevents other tests (potentially providing additional coverage) from being run in that same timeframe. In an effort to address this, in the work we have been working on utilizing SE mode’s avoiding modeling system calls to speed up the runtime of the GPU regression tests. Specifically, we redesign the input reading phase of these GPU tests to create and use mmap’d files for their input arrays (which SE mode completes all at once) instead of reading in the files entry by entry. In doing so, we see significant reductions in runtime of at least 29% 

   more » « less
2. Improving the Speed of gem5’s GPU Regression Tests

   James Braun; Matthew D. Sinclair (June 2023, 5th gem5 Users' Workshop)

   n recent years, we have been enhancing and updating gem5’s GPU support. First, we have enhanced gem5’s GPU support for ML workloads such that gem5 can now run. Moreover, as part of this support, we created, validated, and released a Docker image that contains the proper software and libraries needed to run GCN3 and Vega GPU models in gem5. With this container, users can run the gem5 GPU model, as well as build the ROCm applications that they want to run in the GPU model, out of the box without needing to properly install the appropriate ROCm software and libraries. Additionally, we have updated gem5 to make it easier to reproduce results, including releasing support for a number of GPU workloads in gem5-resources and enabling continuous integration testing on future GPU commits. However, in an effort to provide sufficient coverage, the cur- rent testing support for GPU tests requires significant runtime both for the nightly and weekly regression tests. Currently most of these regression tests test the GPU SE mode support, since GPU FS mode support is still nascent. Unfortunately, much of this time is spent parsing input files to create arrays and other data structures that the GPU subsequently computes on. Although SE mode does not simulate the system calls needed to read these input files, nevertheless this still represents a significant overhead that increases runtime and prevents other tests (potentially providing additional coverage) from being run in that same timeframe. In an effort to address this, in the work we have been working on utilizing SE mode’s avoiding modeling system calls to speed up the runtime of the GPU regression tests. Specifically, we redesign the input reading phase of these GPU tests to create and use mmap’d files for their input arrays (which SE mode completes all at once) instead of reading in the files entry by entry. In doing so, we see significant reductions in runtime of at least 29% 

   more » « less
3. Frontiers: Algorithmic Collusion: Supra-competitive Prices via Independent Algorithms

   https://doi.org/10.1287/mksc.2020.1276  

   Hansen, Karsten T.; Misra, Kanishka; Pai, Mallesh M. (January 2021, Marketing Science)

   null (Ed.)

   Motivated by their increasing prevalence, we study outcomes when competing sellers use machine learning algorithms to run real-time dynamic price experiments. These algorithms are often misspecified, ignoring the effect of factors outside their control, for example, competitors’ prices. We show that the long-run prices depend on the informational value (or signal-to-noise ratio) of price experiments: if low, the long-run prices are consistent with the static Nash equilibrium of the corresponding full information setting. However, if high, the long-run prices are supra-competitive—the full information joint monopoly outcome is possible. We show that this occurs via a novel channel: competitors’ algorithms’ prices end up running correlated experiments. Therefore, sellers’ misspecified models overestimate the own price sensitivity, resulting in higher prices. We discuss the implications on competition policy. 

   more » « less
4. The Tax Elasticity of Capital Gains and Revenue-Maximizing Rates

   https://doi.org/10.1257/aeri.20200535  

   Agersnap, Ole; Zidar, Owen (December 2021, American Economic Review: Insights)

   This paper uses a direct-projections approach to estimate the effect of capital gains taxation on realizations at the state level and then develops a framework for determining revenue-maximizing rates at the federal level. We find that the elasticity of revenues with respect to the tax rate over a 10-year period is −0.5 to −0.3, indicating that capital gains tax cuts do not pay for themselves and that a 5 percentage point rate increase would yield $18 to $30 billion in annual federal tax revenue. Our long-run estimates yield revenue-maximizing capital gains tax rates of 38 to 47 percent. (JEL E62, H25, H71) 

   more » « less
5. The Earnings and Income Mobility Consequences of Attending a Historically Black College/University: Matching Estimates From 2015 U.S. Department of Education College ScorecardData

   Juliet U. Elu, Jared Ireland (October 2019, The Review of Black political economy)

   This paper considers the labor market consequences of attending a Historically Black College/University (HBCU). With 2015 U.S. Department of Education College Scorecard Data, we use a matching estimator to identify and estimate the treatment effect of HBCU attendance on median earnings, earnings relative to a high school graduate, and income relative to that of the household at time of initial enrollment, 6 and 10 years after attendance. Our treatment effect parameter estimates suggest that once we account for the differential return to college majors, the urban wage premium, and the proportionality/dependence of the labor market return of black student college attendees on the share of a college/university's student population that is black, there is a long-run earnings premium associated with HBCU attendance. In addition, for HBCUs in general, we find that there is a population of students who would realize a positive labor market premium---as high as approximately 42 percent---and earn more than a high school graduate if they were to attend an HBCU. With respect to intergenerational income mobility, we find that HBCU attendance enables their actual and potential attendees to move to a higher quantile of income relative to their households in the long-run. 

   more » « less