MétaCan
Menu
Back to cohort

Accelerated pooling

2021· dissertation· en· W4388853272 on OpenAlexfundno aff
Caio Salvador Rohwedder

Bibliographic record

Venuenot available
Typedissertation
Languageen
FieldComputer Science
TopicAdvanced Neural Network Applications
Canadian institutionsnot available
FundersConselho Nacional de Desenvolvimento Científico e TecnológicoCoordenação de Aperfeiçoamento de Pessoal de Nível SuperiorUniversity of Alberta
KeywordsComputer scienceParallel computingPoolingSpeedupOverhead (engineering)Convolution (computer science)BottleneckVectorization (mathematics)Convolutional neural networkComputer engineeringArtificial intelligenceEmbedded systemArtificial neural networkProgramming language

Abstract

fetched live from OpenAlex

Convolution is an operation crucial to Deep Learning applications.As such, it has been the focus of many optimization efforts on this application domain.Image-to-column (Im2col) and column-to-image (Col2im) are data transformations extensively used to map convolution to matrix multiplication.These transformations rearrange the inputs of convolution to avoid its strided memory access pattern, thus providing a friendlier data layout for CPUs and GPUs.In artificial intelligence (AI) accelerators, these transformations allow convolution to be computed in matrix-multiplier units.Implemented in software, however, they impose a significant overhead that must be compensated by the efficiency gains of matrix-multipliers.DaVinci is an AI accelerator architecture that introduces instructions to optimize Im2col and Col2im, thus lowering the overhead of executing convolution in its matrix-multiplier.Another core layer of convolutional neural networks that presents a similar memory access pattern to convolution is pooling.The execution of pooling is typically targeted to vector computational units.Nevertheless, implementations based on the Im2col and Col2im transformations can be leveraged to improve its execution.This work explores the use of Im2col and Col2im instructions of DaVinci to accelerate pooling layers.The proposed approach uses a general-purpose vector computational unit and instructions primarily designed for convolution.An experimental evaluation reveals that the proposed pooling implementations can yield up to 5.8x speedup compared to baseline implementations that do not use these specialized instructions.The speedups follow from an improved memory layout in the inputs of pooling, as this layout leads to better vectorization of its instructions.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.468
Threshold uncertainty score0.601

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.031
GPT teacher head0.312
Teacher spread0.281 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2021
Admission routes1
Has abstractyes

Explore more

Same topicAdvanced Neural Network ApplicationsFrench-language works237,207