MearcAll papers

arXiv 1409.4842 · cs.CV

Going Deeper with Convolutions

Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, and 3 more

Claims were extracted automatically by reducer-1 on 2026-11-12. The paper's license does not allow republishing it, so long claims are shortened to a quotation. Read the full text on arXiv.

No reproduction recordedclm_92d33442
In our case, the word “deep” is used in two different meanings: first of all, in the sense that we introduce a new level of organization in the form of … (excerpt)

Introduction · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_ffb2e856
The benefits of the architecture are experimentally verified on the ILSVRC 2014 classification and detection challenges, on which it significantly outperforms the current state of the art.

Introduction · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_d5292b70
Despite being a highly speculative undertaking, only after two iterations on the exact choice of topology, we could already see modest gains against the reference architecture based on [12].

Motivation and High Level Considerations · page 4 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_7ff1c030
After further tuning of learning rate, hyperparameters and improved training methodology, we established that the resulting Inception architecture was especially useful in the context of localization and object detection as … (excerpt)

Motivation and High Level Considerations · page 4 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_57e83d48
As these “Inception modules” are stacked on top of each other, their output correlation statistics are bound to vary: as features of higher abstraction are captured by higher layers, their … (excerpt)

Architectural Details · page 4 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_c6ece32c
It was found that a move from fully connected layers to average pooling improved the top-1 accuracy by about 0.6%, however the use of dropout remained essential even after removing … (excerpt)

GoogLeNet · page 6 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_724427b4
The overall effect of cutting the number of proposals with increased coverage is a 1% improvement of the mean average precision for the single model case.

ILSVRC 2014 Detection Challenge Setup and Results · page 10 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_0f623343
Finally, we use an ensemble of 6 ConvNets when classifying each region which improves results from 40% to 43.9% accuracy.

ILSVRC 2014 Detection Challenge Setup and Results · page 10 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedres_636230ef
The top performing model is by Deep Insight and surprisingly only improves by 0.3 points with an ensemble of 3 models while the GoogLeNet obtains significantly stronger results with the … (excerpt)

ILSVRC 2014 Detection Challenge Setup and Results · page 10 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_d095e964
Our results seem to yield a solid evidence that approximating the expected optimal sparse structure by readily available dense building blocks is a viable method for improving neural networks for … (excerpt)

Conclusions · page 10 · Automated extraction, not reviewed · 0 reviewed reproductions

No reproduction recordedclm_eaf0cd3d
Although it is expected that similar quality of result can be achieved by much more expensive networks of similar depth and width, our approach yields solid evidence that moving to … (excerpt)

Conclusions · page 11 · Automated extraction, not reviewed · 0 reviewed reproductions

Reproduced one of these claims?

Record your result in the Mearc app on your own machine, then use Publish this result with this paper's id (1409.4842) and the claim id shown next to the claim. Mearc creates a file that holds only what you choose to publish. Add it to the public claim data by pull request; a maintainer other than you reviews it before it counts.