On-device Federated Learning with Flower

Federated Learning (FL) allows edge devices to collaboratively learn a shared prediction model while keeping their training data on the device, thereby decoupling the ability to do machine learning from the need to store data in the cloud. Despite the algorithmic advancements in FL, the support for on-device training of FL algorithms on edge devices remains poor. In this paper, we present an exploration of on-device FL on various smartphones and embedded devices using the Flower framework. We also evaluate the system costs of on-device FL and discuss how this quantification could be used to design more efficient FL algorithms.

Adaptive Filters and Aggregator Fusion for Efficient Graph Convolutions

Training and deploying graph neural networks (GNNs) remains difficult due to their high memory consumption and inference latency. In this work we present a new type of GNN architecture that achieves state-of-the-art performance with lower memory consumption and latency, along with characteristics suited to accelerator implementation. Our proposal uses memory proportional to the number of vertices in the graph, in contrast to competing methods which require memory proportional to the number of edges; we find our efficient approach actually achieves higher accuracy than competing approaches across 5 large and varied datasets against strong baselines. We achieve our results by using a novel adaptive filtering approach inspired by signal processing; it can be interpreted as enabling each vertex to have its own weight matrix, and is not related to attention. Following our focus on efficient hardware usage, we propose aggregator fusion, a technique to enable GNNs to significantly boost their representational power, with only a small increase in latency of 19% over standard sparse matrix multiplication. Code and pretrained models can be found at this URL: https://github.com/shyam196/egc.

unzipFPGA: Enhancing FPGA-based CNN Engines with On-the-Fly Weights Generation

Single computation engines have become a popular design choice for FPGA-based convolutional neural networks (CNNs) enabling the deployment of diverse models without fabric reconfiguration. This flexibility, however, often comes with significantly reduced performance on memory-bound layers and resource underutilisation due to suboptimal mapping of certain layers on the engine's fixed configuration. In this work, we investigate the implications in terms of CNN engine design for a class of models that introduce a pre-convolution stage to decompress the weights at run time. We refer to these approaches as on-the-fly. To minimise the negative impact of limited bandwidth on memory-bound layers, we present a novel hardware component that enables the on-chip on-the-fly generation of weights. We further introduce an input selective processing element (PE) design that balances the load between PEs on suboptimally mapped layers. Finally, we present unzipFPGA, a framework to train on-the-fly models and traverse the design space to select the highest performing CNN engine configuration. Quantitative evaluation shows that unzipFPGA yields an average speedup of 2.14x and 71% over optimised status-quo and pruned CNN engines under constrained bandwidth and up to 3.69x higher performance density over the state-of-the-art FPGA-based CNN accelerators.

FjORD: Fair and Accurate Federated Learning under heterogeneous targets with Ordered Dropout

Federated Learning (FL) has been gaining significant traction across different ML tasks, ranging from vision to keyboard predictions. In large-scale deployments, client heterogeneity is a fact, and constitutes a primary problem for fairness, training performance and accuracy. Although significant efforts have been made into tackling statistical data heterogeneity, the diversity in the processing capabilities and network bandwidth of clients, termed as system heterogeneity, has remained largely unexplored. Current solutions either disregard a large portion of available devices or set a uniform limit on the model's capacity, restricted by the least capable participants. In this work, we introduce Ordered Dropout, a mechanism that achieves an ordered, nested representation of knowledge in Neural Networks and enables the extraction of lower footprint submodels without the need of retraining. We further show that for linear maps our Ordered Dropout is equivalent to SVD. We employ this technique, along with a self-distillation methodology, in the realm of FL in a framework called FjORD. FjORD alleviates the problem of client system heterogeneity by tailoring the model width to the client's capabilities. Extensive evaluation on both CNNs and RNNs across diverse modalities shows that FjORD consistently leads to significant performance gains over state-of-the-art baselines, while maintaining its nested structure.

A first look into the carbon footprint of federated learning

Despite impressive results, deep learning-based technologies also raise severe privacy and environmental concerns induced by the training procedure often conducted in datacenters. In response, alternatives to centralized training such as Federated Learning (FL) have emerged. Perhaps unexpectedly, FL, in particular, is starting to be deployed at a global scale by companies that must adhere to new legal demands and policies originating from governments and civil society for privacy protection. However, the potential environmental impact related to FL remains unclear and unexplored. This paper offers the first-ever systematic study of the carbon footprint of FL. First, we propose a rigorous model to quantify the carbon footprint, hence facilitating the investigation of the relationship between FL design and carbon emissions. Then, we compare the carbon footprint of FL to traditional centralized learning. Our findings show that FL, despite being slower to converge in some cases, may result in a comparatively greener impact than a centralized equivalent setup. We performed extensive experiments across different types of datasets, settings, and various deep learning models with FL. Finally, we highlight and connect the reported results to the future challenges and trends in FL to reduce its environmental impact, including algorithms efficiency, hardware capabilities, and stronger industry transparency.

It's always personal: Using Early Exits for Efficient On-Device CNN Personalisation

On-device machine learning is becoming a reality thanks to the availability of powerful hardware and model compression techniques. Typically, these models are pretrained on large GPU clusters and have enough parameters to generalise across a wide variety of inputs. In this work, we observe that a much smaller, personalised model can be employed to fit a specific scenario, resulting in both higher accuracy and faster execution. Nevertheless, on-device training is extremely challenging, imposing excessive computational and memory requirements even for flagship smartphones. At the same time, on-device data availability might be limited and samples are most frequently unlabelled. To this end, we introduce PersEPhonEE, a framework that attaches early exits on the model and personalises them on-device. These allow the model to progressively bypass a larger part of the computation as more personalised data become available. Moreover, we introduce an efficient on-device algorithm that trains the early exits in a semi-supervised manner at a fraction of the whole network's personalisation time. Results show that PersEPhonEE boosts accuracy by up to 15.9% while dropping the training cost by up to 2.2x and inference latency by 2.2-3.2x on average for the same accuracy, depending on the availability of labels on-device.

Zero-Cost Proxies for Lightweight NAS

Neural Architecture Search (NAS) is quickly becoming the standard methodology to design neural network models. However, NAS is typically compute-intensive because multiple models need to be evaluated before choosing the best one. To reduce the computational power and time needed, a proxy task is often used for evaluating each model instead of full training. In this paper, we evaluate conventional reduced-training proxies and quantify how well they preserve ranking between multiple models during search when compared with the rankings produced by final trained accuracy. We propose a series of zero-cost proxies, based on recent pruning literature, that use just a single minibatch of training data to compute a model's score. Our zero-cost proxies use 3 orders of magnitude less computation but can match and even outperform conventional proxies. For example, Spearman's rank correlation coefficient between final validation accuracy and our best zero-cost proxy on NAS-Bench-201 is 0.82, compared to 0.61 for EcoNAS (a recently proposed reduced-training proxy). Finally, we use these zero-cost proxies to enhance existing NAS search algorithms such as random search, reinforcement learning, evolutionary search and predictor-based search. For all search methodologies and across three different NAS datasets, we are able to significantly improve sample efficiency, and thereby decrease computation, by using our zero-cost proxies. For example on NAS-Bench-101, we achieved the same accuracy 4$\times$ quicker than the best previous result.

Leveraging Activity Recognition to Enable Protective Behavior Detection in Continuous Data

Protective behavior exhibited by people with chronic pain (CP) during physical activities is the key to understanding their physical and emotional states. Existing automatic protective behavior detection (PBD) methods depend on pre-segmentation of activity instances as they expect situations where activity types are predefined. However, during everyday management, people pass from one activity to another, and support should be delivered continuously and personalized to the activity type and presence of protective behavior. Hence, to facilitate ubiquitous CP management, it becomes critical to enable accurate PBD over continuous data. In this paper, we propose to integrate automatic human activity recognition (HAR) with PBD via a novel hierarchical HAR-PBD architecture comprising GC-LSTM networks, and alleviate the class imbalances therein using a CFCC loss function. Through in-depth evaluation of the approach using a CP patients' dataset, we show that the leveraging of HAR, GC-LSTM networks and the CFCC loss function leads to clear increase in PBD performance against the state-of-the-art (macro F1 score of 0.81 vs. 0.66 and PR-AUC of 0.60 vs. 0.44). We conclude by discussing possible use cases of the HAR-PBD architecture in the context of CP management and other situations. We also discuss the current limitations and ways forward.