ترغب بنشر مسار تعليمي؟ اضغط هنا

Accelerator-aware Neural Network Design using AutoML

112   0   0.0 ( 0 )
 نشر من قبل Suyog Gupta
 تاريخ النشر 2020
والبحث باللغة English




اسأل ChatGPT حول البحث

While neural network hardware accelerators provide a substantial amount of raw compute throughput, the models deployed on them must be co-designed for the underlying hardware architecture to obtain the optimal system performance. We present a class of computer vision models designed using hardware-aware neural architecture search and customized to run on the Edge TPU, Googles neural network hardware accelerator for low-power, edge devices. For the Edge TPU in Coral devices, these models enable real-time image classification performance while achieving accuracy typically seen only with larger, compute-heavy models running in data centers. On Pixel 4s Edge TPU, these models improve the accuracy-latency tradeoff over existing SoTA mobile models.



قيم البحث

اقرأ أيضاً

Majority of the existing graph neural networks (GNN) learn node embeddings that encode their local neighborhoods but not their positions. Consequently, two nodes that are vastly distant but located in similar local neighborhoods map to similar embedd ings in those networks. This limitation prevents accurate performance in predictive tasks that rely on position information. In this paper, we develop GraphReach, a position-aware inductive GNN that captures the global positions of nodes through reachability estimations with respect to a set of anchor nodes. The anchors are strategically selected so that reachability estimations across all the nodes are maximized. We show that this combinatorial anchor selection problem is NP-hard and, consequently, develop a greedy (1-1/e) approximation heuristic. Empirical evaluation against state-of-the-art GNN architectures reveal that GraphReach provides up to 40% relative improvement in accuracy. In addition, it is more robust to adversarial attacks.
398 - Daniel Jung 2019
Localization of unknown faults in industrial systems is a difficult task for data-driven diagnosis methods. The classification performance of many machine learning methods relies on the quality of training data. Unknown faults, for example faults not represented in training data, can be detected using, for example, anomaly classifiers. However, mapping these unknown faults to an actual location in the real system is a non-trivial problem. In model-based diagnosis, physical-based models are used to create residuals that isolate faults by mapping model equations to faulty system components. Developing sufficiently accurate physical-based models can be a time-consuming process. Hybrid modeling methods combining physical-based methods and machine learning is one solution to design data-driven residuals for fault isolation. In this work, a set of neural network-based residuals are designed by incorporating physical insights about the system behavior in the residual model structure. The residuals are trained using only fault-free data and a simulation case study shows that they can be used to perform fault isolation and localization of unknown faults in the system.
Ischemic heart disease (IHD), particularly in its chronic stable form, is a subtle pathology due to its silent behavior before developing in unstable angina, myocardial infarction or sudden cardiac death. Machine learning techniques applied to parame ters extracted form heart rate variability (HRV) signal seem to be a valuable support in the early diagnosis of some cardiac diseases. However, so far, IHD patients were identified using Artificial Neural Networks (ANNs) applied to a limited number of HRV parameters and only to very few subjects. In this study, we used several linear and non-linear HRV parameters applied to ANNs, in order to confirm these results on a large cohort of 965 sample of subjects and to identify which features could discriminate IHD patients with high accuracy. By using principal component analysis and stepwise regression, we reduced the original 17 parameters to five, used as inputs, for a series of ANNs. The highest accuracy of 82% was achieved using meanRR, LFn, SD1, gender and age parameters and two hidden neurons.
Real-time state estimation of dynamical systems is a fundamental task in signal processing and control. For systems that are well-represented by a fully known linear Gaussian state space (SS) model, the celebrated Kalman filter (KF) is a low complexi ty optimal solution. However, both linearity of the underlying SS model and accurate knowledge of it are often not encountered in practice. Here, we present KalmanNet, a real-time state estimator that learns from data to carry out Kalman filtering under non-linear dynamics with partial information. By incorporating the structural SS model with a dedicated recurrent neural network module in the flow of the KF, we retain data efficiency and interpretability of the classic algorithm while implicitly learning complex dynamics from data. We numerically demonstrate that KalmanNet overcomes nonlinearities and model mismatch, outperforming classic filtering methods operating with both mismatched and accurate domain knowledge.
Designing an incentive compatible auction that maximizes expected revenue is a central problem in Auction Design. Theoretical approaches to the problem have hit some limits in the past decades and analytical solutions are known for only a few simple settings. Computational approaches to the problem through the use of LPs have their own set of limitations. Building on the success of deep learning, a new approach was recently proposed by Duetting et al. (2019) in which the auction is modeled by a feed-forward neural network and the design problem is framed as a learning problem. The neural architectures used in that work are general purpose and do not take advantage of any of the symmetries the problem could present, such as permutation equivariance. In this work, we consider auction design problems that have permutation-equivariant symmetry and construct a neural architecture that is capable of perfectly recovering the permutation-equivariant optimal mechanism, which we show is not possible with the previous architecture. We demonstrate that permutation-equivariant architectures are not only capable of recovering previous results, they also have better generalization properties.

الأسئلة المقترحة

التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا