深度學(xué)習(xí)概述

I have found the DeepSense framework as one of the promising deep learning architectures for processing Time-Series sensing data. In this brief and intuitive overview, I’ll present the main ideas of the original paper titled “Deep Sense: A Unified DL Framework for Time-Series Mobile Sensing Data Processing” by Yao et-al. (Can be found at www2017).

我發(fā)現(xiàn)DeepSense框架是用于處理時間序列感應(yīng)數(shù)據(jù)的有前途的深度學(xué)習(xí)架構(gòu)之一。 在這個簡短而直觀的概述中，我將介紹Yao等人題為“深度感知：用于時間序列移動傳感數(shù)據(jù)處理的統(tǒng)一DL框架”的原始論文的主要思想。 (可以在www2017上找到)。

主要問題 (The Main Problem)

DeepSense addresses the problem of mobile sensing using time-series data from various sensors (accelerometers, gyroscopes, and more). Due to the quality of the sensor, processing any type of sensor involves noisy measurements. It is difficult to find a distribution that describes noise in real-life, so in the end, our measurements are corrupted by ”unknown” noise (nonlinear, correlated over time, etc.). In addition, for tracking applications, a physical system model should be defined. This model is also affected by “noise” in terms of inaccuracy (as we cannot predict/ define the model without error). As we deal with real-time measurement (and physical model), the behavior can be described as “time-series measurements of physical quantities..”.

DeepSense使用來自各種傳感器(加速度計，陀螺儀等)的時間序列數(shù)據(jù)解決了移動感應(yīng)的問題。 由于傳感器的質(zhì)量，處理任何類型的傳感器都會涉及噪聲測量。 很難找到描述現(xiàn)實生活中的噪聲的分布，因此最后，我們的測量結(jié)果被“未知”噪聲(非線性，隨時間相關(guān)等)破壞了。 此外，對于跟蹤應(yīng)用程序，應(yīng)定義一個物理系統(tǒng)模型。 該模型在準(zhǔn)確性方面也受到“噪聲”的影響(因為我們無法無誤地預(yù)測/定義模型)。 當(dāng)我們處理實時測量(和物理模型)時，該行為可以描述為“物理量的時間序列測量”。

(核心)想法 (The (core) Idea)

The author proposes a unified DL framework that addresses the challenges described in mobile sensing: integration of Convolution Neural Networks (CNN) and Recurrent Neural Network (RNN). CNN is responsible for the computation of the sensing quantities within the time interval where it extracts the local features for all sensors and combines them into global features. RNN is responsible for computation of the sensing quantities across time intervals, where it extracts temporal dependencies. The DeepSense solves both classification and regression of mobile computing tasks in some unified manner.

作者提出了一個統(tǒng)一的DL框架，以解決移動傳感中描述的挑戰(zhàn)：卷積神經(jīng)網(wǎng)絡(luò)(CNN)和遞歸神經(jīng)網(wǎng)絡(luò)(RNN)的集成。 CNN負(fù)責(zé)計算時間間隔內(nèi)的感應(yīng)量， 在該時間間隔內(nèi) ，CNN提取所有傳感器的局部特征并將其組合為全局特征。 RNN負(fù)責(zé)跨時間間隔傳感量的計算，在那里它提取時間相關(guān)。 DeepSense以某種統(tǒng)一的方式解決了移動計算任務(wù)的分類和回歸問題。

架構(gòu) (The Architecture)

The architecture consists of 3 parts: a convolutional layer, recurrent layer, and output layer. Since DeepSense can be used for classification and regression tasks, the output layer should be set according to the specific task.

該體系結(jié)構(gòu)由三部分組成：卷積層，循環(huán)層和輸出層。 由于DeepSense可用于分類和回歸任務(wù)，因此應(yīng)根據(jù)特定任務(wù)設(shè)置輸出層。

Convolutional layer: an individual convolutional subnet can be represented by 3 layers (Individual Convolutional Layers 1,2,3). By applying a 2d filter, the net can learn interaction among sensor measurements and local patterns. Next, by Flatten & Concatenation layer, the matrices are flattened into a vector which is the input to the next 3 layers: The Merge Convolutional Layers. A 2d filter is also applied here to learn the interaction between all K input sensors. For each convolutional layer, DeepSense learns 64 filters by using ReLU. Apply batch normalization at each layer to reduce the internal covariate shift. A Flatten & Concatenation layer completes this phase of the convolutional layer.

卷積層：單個卷積子網(wǎng)可以由3層表示( 單個卷積層 1,2,3)。 通過應(yīng)用二維過濾器，網(wǎng)絡(luò)可以了解傳感器測量值和局部模式之間的相互作用。 接下來，通過Flatten＆Concatenation層，將矩陣展平為一個向量，該向量是接下來3層的輸入： 合并卷積層。 這里還應(yīng)用了2d濾波器，以了解所有K個輸入傳感器之間的相互作用。 對于每個卷積層，DeepSense通過使用ReLU學(xué)習(xí)64個過濾器。 在每一層應(yīng)用批量歸一化以減少內(nèi)部協(xié)變量偏移。 Flatten＆Concatenation層完成了卷積層的此階段。

Recurrent layer: the power of RNN is in the ability to approximate function and understand the important features for time-series. In this architecture, the RNN extension model that used is a stacked Gated Recurrent Unit (GRU). 2 layers of RNN are implemented with dropout to the connection between these Recurrent Layers (1,2), as also recurrent batch normalization in order to reduce the internal covariance shift among data series.

循環(huán)層： RNN的功能在于能夠逼近函數(shù)并了解時間序列的重要特征。 在此體系結(jié)構(gòu)中，使用的RNN擴(kuò)展模型是堆疊的門控循環(huán)單元(GRU)。 RNN的2層實現(xiàn)為與這些遞歸層(1,2)之間的連接斷開，也為遞歸批歸一化，以減少數(shù)據(jù)序列之間的內(nèi)部協(xié)方差漂移。

Output Layer: Up to this phase, the net has a series of vectors for each time step. Now, we should handle carefully upon the tasks: regression or classification. For regression, a dictionary should be learned. For classification, the composition of the output layer should be done by averaging the features over time. then, feeding the final features into a softmax layer to generate the prediction.

輸出層：到此階段，網(wǎng)絡(luò)每個時間步都有一系列矢量。 現(xiàn)在，我們應(yīng)該認(rèn)真處理以下任務(wù)：回歸或分類。 為了回歸，應(yīng)該學(xué)習(xí)字典。 對于分類，輸出層的組成應(yīng)通過對特征隨時間進(jìn)行平均來完成。 然后，將最終特征輸入softmax層以生成預(yù)測。

摘要和我的觀點 (Summary and My Point of View)

DeepSense seems to be very promising for many time-series tasks, Aside the Temporal Convolutional Networks, WaveNet, and others. When I first read the paper, I thought about the Extended Kalman Filter as a nice classical compression — as it deals with the same type of problem. This deep learning architecture might deal with model uncertainty, noisy measurement, and more.

除了時間卷積網(wǎng)絡(luò)，WaveNet等，DeepSense對于許多時間序列任務(wù)似乎都非常有前途。 當(dāng)我第一次閱讀本文時，我認(rèn)為擴(kuò)展卡爾曼濾波器是一種很好的經(jīng)典壓縮方法，因為它可以處理相同類型的問題。 這種深度學(xué)習(xí)架構(gòu)可能會處理模型不確定性，噪聲測量等問題。

Further reading

進(jìn)一步閱讀

Yao, Shuochao, et al. “Deepsense: A unified deep learning framework for time-series mobile sensing data processing.” Proceedings of the 26th International Conference on World Wide Web. 2017.

姚碩超，等。 “ Deepsense：用于時間序列移動感測數(shù)據(jù)處理的統(tǒng)一深度學(xué)習(xí)框架。” 第26屆國際萬維網(wǎng)會議論文集 。 2017。

翻譯自: https://towardsdatascience.com/an-overview-of-deepsense-framework-5d7704343734

深度學(xué)習(xí)概述

總結(jié)

以上是生活随笔為你收集整理的深度学习概述_深度感测框架概述的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。