The recent groundbreaking advances in deep learning networks (DNNs) make them attractive for embedded systems. However, it can take a long time for DNNs to make an inference on resource-limited embedded devices. Offloading the computation into the cloud is often infeasible due to privacy concerns, high latency, or the lack of connectivity. As such, there is a critical need to find a way to effectively execute the DNN models locally on the devices. This paper presents an adaptive scheme to determine which DNN model to use for a given input, by considering the desired accuracy and inference time. Our approach employs machine learning to develop a predictive model to quickly select a pre-trained DNN to use for a given input and the optimization constraint. We achieve this by first training off-line a predictive model, and then use the learnt model to select a DNN model to use for new, unseen inputs. We apply our approach to the image classification task and evaluate it on a Jetson TX2 embedded deep learning platform using the ImageNet ILSVRC 2012 validation dataset. We consider a range of influential DNN models. Experimental results show that our approach achieves a 7.52% improvement in inference accuracy, and a 1.8x reduction in inference time over the most-capable, single DNN model.

The recent ground-breaking advances in deep learning networks (DNNs) make them attractive for embedded systems.However, it can take a long time for DNNs to make an inference on resource-limited embedded devices. Offloading the computation into the cloud is often infeasible due to privacy concerns, high latency, or the lack of connectivity. As such, there is a critical need to find a way to effectively execute the DNN models locally on the devices. This paper presents an adaptive scheme to determine which DNN model to use for a given input, by considering the desired accuracy and inference time. Our approach employs machine learning to develop a predictive model to quickly select a pre-trained DNN to use for a given input and the optimization constraint. We achieve this by first training off-line a predictive model, and then use the learnt model to select a DNN model to use for new, unseen inputs. We apply our approach to the image classification task and evaluate it on a Jetson TX2 embedded deep learning platform using the ImageNet ILSVRC 2012 validation dataset. We consider a range of influential DNN models. Experimental results show that our approach achieves a 7.52% improvement in inference accuracy, and a 1.8x reduction in inference time over the most-capable single DNN model.

Pattern lock is widely used for identiication and authentication on Android devices. This article presents a novel video-based side channel attack that can reconstruct Android locking patterns from video footage ilmed using a smartphone. As a departure from previous attacks on pattern lock, this new attack does not require the camera to capture any content displayed on the screen. Instead, it employs a computer vision algorithm to track the ingertip movement trajectory to infer the pattern. Using the geometry information extracted from the tracked ingertip motions, the method can accurately infer a small number of (often one) candidate patterns to be tested by an attacker. We conduct extensive experiments to evaluate our approach using 120 unique patterns collected from 215 independent users. Experimental results show that the proposed attack can reconstruct over 95% of the patterns in ive attempts. We discovered that, in contrast to most people's belief, complex patterns do not ofer stronger protection under our attacking scenarios. This is demonstrated by the fact that we are able to break all but one complex patterns (with a 97.5% success rate) as opposed to 60% of the simple patterns in the irst attempt. We demonstrate that this video-side channel is a serious concern for not only graphical locking patterns but also PIN-based passwords, as algorithms and analysis developed from the attack can be easily adapted to target PIN-based passwords. As a countermeasure, we propose to change the way the Android locking pattern is constructed and used. We show that our proposal can successfully defeat this video-based attack. We hope the results of this article can encourage the community to revisit the design and practical use of Android pattern lock.

Summary In this paper, we present a new runtime code generation technique for speculative loop optimization and parallelization. The main benefit of this technique, compared to previous approaches, is to enable advanced optimizing loop transformations at runtime with an acceptable time overhead. The loop transformations that may be applied are those handled by the polyhedral model. The proposed code generation strategy is based on the generation of “code‐bones” at compile‐time, which are parametrized code snippets either dedicated to speculation management or to computations of the original target program. These code‐bones are then instantiated and assembled at runtime to constitute the speculatively optimized code, as soon as an optimizing polyhedral transformation has been determined. Their granularity threshold is sufficient to apply any polyhedral transformation, while still enabling fast runtime code generation. This approach has been implemented in the speculative loop parallelizing framework APOLLO.