Vera CPU

X. Vera

B. Vera, José Carmona-Márquez, C. Vidal-Giné, F. Fernández-Calderón

Abstract Introduction Although simultaneous polysubstance use (SPU) is associated with greater harms than concurrent polysubstance use (CPU), no previous research has examined the effectiveness of harm reduction strategies in reducing drug-related harms in people who use drugs simultaneously and whether these strategies have a differential impact on drug-related negative consequences according to SPU patterns. Objectives: to examine the relationship between SPU patterns, harm reduction strategies and drug-related consequences experienced among people who attend dance music settings, and to examine the moderating role of SPU patterns in the relationship between harm reduction strategies and drug-related negative consequences Methods: a web-based survey was used to gather data from 649 substance-using attendees at dance music settings (mean age = 26.2, 68% male). The survey collected data on: settings of last party attended, drug use, harm reduction strategies used and drug-related negative consequences experienced during last party attended. Results: Latent class analysis identified two SPU profiles: Moderate SPU and Severe SPU. Participants in the severe SPU class experienced more drug-related negative consequences at their last party in comparison with those in the moderate SPU class. Regardless of SPU pattern, avoiding binge drinking was related to fewer drug-related negative consequences, whilst avoiding mixing stimulants was related to fewer drug-related negative consequences only among those participants in the severe simultaneous polysubstance use class. Conclusions: These findings could help to inform the design of messages and tailored interventions aimed at minimizing drug-related negative consequences among people who use multiple substances in the dance music scene.

Chase Cummins, Richard Veras

Register allocation is one of the most important problems for modern compilers. With a practically unlimited number of user variables and a small number of CPU registers, assigning variables to registers without conflicts is a complex task. This work demonstrates the use of casting the register allocation problem as a graph coloring problem. Using technologies such as PyTorch and OpenAI Gymnasium Environments we will show that a Proximal Policy Optimization model can learn to solve the graph coloring problem. We will also show that the labeling of a graph is critical to the performance of the model by taking the matrix representation of a graph and permuting it. We then test the model's effectiveness on each of these permutations and show that it is not effective when given a relabeling of the same graph. Our main contribution lies in showing the need for label reordering invariant representations of graphs for machine learning models to achieve consistent performance.

Sudhindra Gopal Krishna, Aditya Narasimhan, S. Radhakrishnan, R. Veras

Matrix multiplication is an essential operation in the field of mathematics and computer science. Many critical computations, such as matrix factorization and graph computations, cast the bulk of their computation in terms of this operation. Thus, it is crucial that this operation is tuned to the data being computed on. In the case of sparse domains, this translates to minimizing the traffic between the CPU and main memory as the amount of work is not necessarily sufficient to amortize the code of the data movement. The amount of memory required to store a nonnegative valued matrix of n rows and m columns requires (n × m) × log2(n) bits. When these dimensions are converted to real world scenarios, for example, a one billion by one billion matrix will require 1000 petabytes of memory, which is impractical. This hinders the ability to perform any operations on the matrix.In this paper, we propose techniques for performing Matrix-Matrix multiplication directly on compressed data stored in two different compression data structures. The structures we consider are the well-known compressed sparse matrix and the Compressed Binary Trees [1]. We test our algorithm on extremely large matrices, in the order of 100s of millions with various levels of sparsity. We show for matrices of order 100 million with 10 million nonzero elements, the space required to store the matrices using the CBT representation is about 6.4MB and requires 13.52s to complete the multiplication using the sequential algorithms provided in this paper.

Fauzi Dwi Setiawan Sumadi, Christian Sri Kusuma Aditya, A. Maulana, Syaifuddin, Vera Suryani