Computational power is crucial for the development and deployment of artificial intelligence capabilities, as the large size of deep learning models often requires significant resources. Compression methods aim to reduce model size making artificial intelligence more sustainable and accessible. Compression techniques are often applied uniformly across model layers, without considering their individual characteristics. In this paper, we introduce a customized approach that optimizes compression for each layer individually. Some layers undergo both pruning and/or quantization, while others are only quantized, with fuzzy logic guiding these decisions. The quantization precision is further adjusted based on the importance of each layer. Our experiments on both supervised and self-supervised models using the librispeech dataset show only a slight decrease in performance, with about 85% memory footprint reduction.

The heterogeneity induced by the federated edge learning execution environment poses many performance challenges. Indeed, a balance between efficient resource usage and inference accuracy must be found. Our work therefore aims at characterizing the hyperparameter influence by creating a variety of simulated execution circumstances. We designed an experimentation platform to simulate the execution of a typical image recognition training workload to highlight tweaking opportunities. We particularly focus on participant selection as an important performance lever. Thus, our benchmarks vary the number of clients participating in the federated edge learning process within i.i.d. and non-i.i.d. environments, while illustrating real-world configurations based on heterogeneous edge systems. We identify computational efficiency facets in federated edge learning and propose a taxonomic methodology to approach the study. We demonstrate the impact of the number of clients selected to participate in the global model update of federated edge learning on the overall system computational efficiency in challenging environments. Thus, we propose an optimization formula to meet computational efficiency and accurate models in challenging federated edge learning environments.

It is often hard to predict the performance of a statically generated code. Hardware availability, hardware specification and problem size may change from one execution context to another. The main contribution of this work is an entirely automatic method aiming to predict execution times of semantically equivalent versions of affine loop nests on GPUs, then, to run the best performing one on GPU or CPU. To make accurate predictions, our framework relies on three consecutive stages: a static code generation, an offline profiling and an online prediction. Different versions are statically generated by PPCG, a source-to-source polyhedral compiler, able to generate CUDA code from static control loops written in C. The code versions differ by their block sizes, tiling and parallel schedule. The profiling code carries out the required measurements on the target machine: throughput between host and device memory, and execution time of the kernels with various parameters. At runtime, we rely on those results to calculate a predicted execution time on GPU. This is followed by a "fastest wins" algorithm, that runs instances of the target code concurrently on CPU and GPU, the first completed kills the other one. We validate this proposal on the polyhedral benchmark suite, showing that the predictions are accurate and that the runtime selection is effective on two different architectures.