深度学习卷积网络：Inception架构解析

5星 · 超过95%的资源需积分: 15 128 浏览量更新于2024-09-11 收藏 1.16MB PDF 举报

"Going deeper with convolutions" 这篇论文“Going Deeper with Convolutions”由Google Inc.的研究人员Christian Szegedy等人撰写，是深度学习领域的一篇重要文献，特别是在计算机视觉中的图像分类和检测上取得了突破性进展。该研究在2014年ImageNet Large-Scale Visual Recognition Challenge (ILSVRC14)中设立了新的标准。论文的主要贡献是一种名为Inception的深度卷积神经网络架构。Inception网络设计的核心目标是更有效地利用网络内部的计算资源。通过精心设计的结构，Inception网络能够在保持计算成本不变的情况下增加网络的深度和宽度。这种设计不仅提高了模型的复杂性，还确保了效率，避免了计算资源的浪费。为了优化网络质量，Inception架构的设计基于赫布定律（Hebbian principle）以及多尺度处理的直觉。赫布定律是一种学习理论，它表明神经元之间的连接强度会因共同激活而增强，这在构建深度学习模型时具有指导意义。多尺度处理则允许模型同时处理不同大小和形状的特征，这对于图像识别等任务至关重要。 Inception网络的一个具体实现版本被称为GoogLeNet，这是一个包含22层的深度网络。在ILSVRC14挑战中，GoogLeNet展示了其卓越的性能，并在图像分类和检测任务上取得了优异的结果。论文详细评估了GoogLeNet在不同环境下的表现，证明了其深度学习模型的有效性。 “Going Deeper with Convolutions”论文推动了深度学习领域的发展，特别是对于卷积神经网络的设计和优化，使得模型能够处理更复杂的任务，同时保持高效运行。Inception架构和GoogLeNet网络至今仍然是许多现代深度学习应用的基础，对后续的卷积网络设计产生了深远影响。

(a) Siberian husky (b) Eskimo dog

Figure 1: Two distinct classes from the 1000 classes of the ILSVRC 2014 classiﬁcation challenge.

and expensive, especially if expert human raters are necessary to distinguish between ﬁne-grained

visual categories like those in ImageNet (even in the 1000-class ILSVRC subset) as demonstrated

by Figure 1.

Another drawback of uniformly increased network size is the dramatically increased use of compu-

tational resources. For example, in a deep vision network, if two convolutional layers are chained,

any uniform increase in the number of their ﬁlters results in a quadratic increase of computation. If

the added capacity is used inefﬁciently (for example, if most weights end up to be close to zero),

then a lot of computation is wasted. Since in practice the computational budget is always ﬁnite, an

efﬁcient distribution of computing resources is preferred to an indiscriminate increase of size, even

when the main objective is to increase the quality of results.

The fundamental way of solving both issues would be by ultimately moving from fully connected

to sparsely connected architectures, even inside the convolutions. Besides mimicking biological

systems, this would also have the advantage of ﬁrmer theoretical underpinnings due to the ground-

breaking work of Arora et al. [2]. Their main result states that if the probability distribution of

the data-set is representable by a large, very sparse deep neural network, then the optimal network

topology can be constructed layer by layer by analyzing the correlation statistics of the activations

of the last layer and clustering neurons with highly correlated outputs. Although the strict math-

ematical proof requires very strong conditions, the fact that this statement resonates with the well

known Hebbian principle – neurons that ﬁre together, wire together – suggests that the underlying

idea is applicable even under less strict conditions, in practice.

On the downside, todays computing infrastructures are very inefﬁcient when it comes to numerical

calculation on non-uniform sparse data structures. Even if the number of arithmetic operations is

reduced by 100×, the overhead of lookups and cache misses is so dominant that switching to sparse

matrices would not pay off. The gap is widened even further by the use of steadily improving,

highly tuned, numerical libraries that allow for extremely fast dense matrix multiplication, exploit-

ing the minute details of the underlying CPU or GPU hardware [16, 9]. Also, non-uniform sparse

models require more sophisticated engineering and computing infrastructure. Most current vision

oriented machine learning systems utilize sparsity in the spatial domain just by the virtue of em-

ploying convolutions. However, convolutions are implemented as collections of dense connections

to the patches in the earlier layer. ConvNets have traditionally used random and sparse connection

tables in the feature dimensions since [11] in order to break the symmetry and improve learning, the

trend changed back to full connections with [9] in order to better optimize parallel computing. The

uniformity of the structure and a large number of ﬁlters and greater batch size allow for utilizing

efﬁcient dense computation.

This raises the question whether there is any hope for a next, intermediate step: an architecture

that makes use of the extra sparsity, even at ﬁlter level, as suggested by the theory, but exploits our

剩余11页未读，继续阅读

xf__mao

粉丝: 843
资源: 10

深度学习卷积网络：Inception架构解析

Going deeper with convolutions论文PPT

【6】Going deeper with convolutions.pdf

Going Deeper with Convolutions

Going deeper with convolutions.xmind

论文精解：Going Deeper with Convolutions

(175797816)华南理工大学信号与系统Signal and Systems期末考试试卷及答案

深圳建设施工项目安全生产奖惩管理制度.docx

离散数学课后题答案+sdut往年试卷+复习提纲资料

自考04741计算机网络原理真题及答案及课件

C++实现rpc，全程手写

最新资源