اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدDeepRecSys: A System for Optimizing End-To-End At-scale Neural Recommendation Inference
92
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Neural personalized recommendation is the corner-stone of a wide collection of cloud services and products, constituting significant compute demand of the cloud infrastructure. Thus, improving the execution efficiency of neural recommendation directly translates into infrastructure capacity saving. In this paper, we devise a novel end-to-end modeling infrastructure, DeepRecInfra, that adopts an algorithm and system co-design methodology to custom-design systems for recommendation use cases. Leveraging the insights from the recommendation characterization, a new dynamic scheduler, DeepRecSched, is proposed to maximize latency-bounded throughput by taking into account characteristics of inference query size and arrival patterns, recommendation model architectures, and underlying hardware systems. By doing so, system throughput is doubled across the eight industry-representative recommendation models. Finally, design, deployment, and evaluation in at-scale production datacenter shows over 30% latency reduction across a wide variety of recommendation models running on hundreds of machines.
قيم البحث
اقرأ أيضاً
In this paper we present a system for monitoring and controlling dynamic network circuits inside the USLHCNet network. This distributed service system provides in near real-time complete topological information for all the circuits, resource allocati
on and usage, accounting, detects automatically failures in the links and network equipment, generate alarms and has the functionality to take automatic actions. The system is developed based on the MonALISA framework, which provides a robust monitoring and controlling service oriented architecture, with no single points of failure.
We present Deep Speaker, a neural speaker embedding system that maps utterances to a hypersphere where speaker similarity is measured by cosine similarity. The embeddings generated by Deep Speaker can be used for many tasks, including speaker identif
ication, verification, and clustering. We experiment with ResCNN and GRU architectures to extract the acoustic features, then mean pool to produce utterance-level speaker embeddings, and train using triplet loss based on cosine similarity. Experiments on three distinct datasets suggest that Deep Speaker outperforms a DNN-based i-vector baseline. For example, Deep Speaker reduces the verification equal error rate by 50% (relatively) and improves the identification accuracy by 60% (relatively) on a text-independent dataset. We also present results that suggest adapting from a model trained with Mandarin can improve accuracy for English speaker recognition.
A typical audio signal processing pipeline includes multiple disjoint analysis stages, including calculation of a time-frequency representation followed by spectrogram-based feature analysis. We show how time-frequency analysis and nonnegative matrix
factorisation can be jointly formulated as a spectral mixture Gaussian process model with nonstationary priors over the amplitude variance parameters. Further, we formulate this nonlinear models state space representation, making it amenable to infinite-horizon Gaussian process regression with approximate inference via expectation propagation, which scales linearly in the number of time steps and quadratically in the state dimensionality. By doing so, we are able to process audio signals with hundreds of thousands of data points. We demonstrate, on various tasks with empirical data, how this inference scheme outperforms more standard techniques that rely on extended Kalman filtering.
The Internet of Things (IoT) promises to help solve a wide range of issues that relate to our wellbeing within application domains that include smart cities, healthcare monitoring, and environmental monitoring. IoT is bringing new wireless sensor use
cases by taking advantage of the computing power and flexibility provided by Edge and Cloud Computing. However, the software and hardware resources used within such applications must perform correctly and optimally. Especially in applications where a failure of resources can be critical. Service Level Agreements (SLA) where the performance requirements of such applications are defined, need to be specified in a standard way that reflects the end-to-end nature of IoT application domains, accounting for the Quality of Service (QoS) metrics within every layer including the Edge, Network Gateways, and Cloud. In this paper, we propose a conceptual model that captures the key entities of an SLA and their relationships, as a prior step for end-to-end SLA specification and composition. Service level objective (SLO) terms are also considered to express the QoS constraints. Moreover, we propose a new SLA grammar which considers workflow activities and the multi-layered nature of IoT applications. Accordingly, we develop a tool for SLA specification and composition that can be used as a template to generate SLAs in a machine-readable format. We demonstrate the effectiveness of the proposed specification language through a literature survey that includes an SLA language comparison analysis, and via reflecting the user satisfaction results of a usability study.
In this paper, we present an end-to-end training framework for building state-of-the-art end-to-end speech recognition systems. Our training system utilizes a cluster of Central Processing Units(CPUs) and Graphics Processing Units (GPUs). The entire
data reading, large scale data augmentation, neural network parameter updates are all performed on-the-fly. We use vocal tract length perturbation [1] and an acoustic simulator [2] for data augmentation. The processed features and labels are sent to the GPU cluster. The Horovod allreduce approach is employed to train neural network parameters. We evaluated the effectiveness of our system on the standard Librispeech corpus [3] and the 10,000-hr anonymized Bixby English dataset. Our end-to-end speech recognition system built using this training infrastructure showed a 2.44 % WER on test-clean of the LibriSpeech test set after applying shallow fusion with a Transformer language model (LM). For the proprietary English Bixby open domain test set, we obtained a WER of 7.92 % using a Bidirectional Full Attention (BFA) end-to-end model after applying shallow fusion with an RNN-LM. When the monotonic chunckwise attention (MoCha) based approach is employed for streaming speech recognition, we obtained a WER of 9.95 % on the same Bixby open domain test set.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
