The network
What rests on what, drawn. Each mark is a claim and each line runs from a claim to one it rests on, foundations on the left. Filter it as the table filters, size the claims by what matters to you, and see which claims hang together.
Around: Using FPN in a basic Faster R-CNN system, our m…
Centred on one claim: 15 claims within its whole group of it.
● established◐ supported○ unchecked◆ contested✕ refuted⊘ tried, not checkable
human literature published here declared by its author identified in the literature refutesleft to right: what rests on what
size: stakes, by area; the largest here 19.1 the claim it is drawn around
The drawing is wider than this screen: drag it sideways to see the rest, or read the table.
Every claim drawn, as a table
| Claim | Status | Checkable | Credence | Use | Stakes | Rests on |
|---|---|---|---|---|---|---|
| On the opposite end in which accuracy is critical, we present a detector that achieves state-of-the-art performance mea… | ○ unchecked | yes | 0.55 | 0 | 11.4 | We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. |
| Our proposed "DeepLab" system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reachi… | ○ unchecked | yes | 0.55 | 0 | 14.4 | We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. |
| We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over… | ○ unchecked | yes | 0.55 | 0 | 9.1 | Batch Normalization allows us to use much higher learning rates and be less careful about initialization. |
| We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, exceed the state-of-the-art in… | ○ unchecked | yes | 0.55 | 0 | 11.4 | We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. |
| Constructed entirely from standard ConvNet modules, ConvNeXts compete favorably with Transformers in terms of accuracy… | ○ unchecked | yes | 0.55 | 0 | 13.0 | On the ImageNet-1K dataset, we empirically show that even under the restricted condition of maintaining complexity, inc… |
| Experiments on ImageNet classification and MS COCO object detection demonstrate the superior performance of ShuffleNet… | ○ unchecked | yes | 0.55 | 0 | 9.8 | On the ImageNet-1K dataset, we empirically show that even under the restricted condition of maintaining complexity, inc…, Batch Normalization allows us to use much higher learning rates and be less careful about initialization., For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while ach… |
| ION improves state-of-the-art on PASCAL VOC 2012 object detection from 73.9% to 76.4% mAP. | ○ unchecked | yes | 0.55 | 0 | 10.3 | We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. |
| Using FPN in a basic Faster R-CNN system, our method achieves state-of-the-art single-model results on the COCO detecti… | ○ unchecked | yes | 0.55 | 0 | 8.8 | For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while ach… |
| On the ImageNet-1K dataset, we empirically show that even under the restricted condition of maintaining complexity, inc… | ○ unchecked | yes | 0.55 | 0 | 10.6 | — |
| We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. | ○ unchecked | yes | 0.55 | 0 | 19.1 | — |
| We show improvements on several image classification and object detection tasks, and report the highest ImageNet-1k sin… | ○ unchecked | yes | 0.55 | 0 | 10.2 | On the ImageNet-1K dataset, we empirically show that even under the restricted condition of maintaining complexity, inc…, Specifically, we show no loss of accuracy when training with large minibatch sizes up to 8192 images. |
| Finally, the learned features by NASNet used with the Faster-RCNN framework surpass state-of-the-art by 4.0% achieving… | ○ unchecked | yes | 0.55 | 0 | 12.7 | For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while ach… |
| Batch Normalization allows us to use much higher learning rates and be less careful about initialization. | ○ unchecked | yes | 0.55 | 0 | 16.1 | — |
| Specifically, we show no loss of accuracy when training with large minibatch sizes up to 8192 images. | ○ unchecked | yes | 0.55 | 0 | 12.2 | — |
| For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while ach… | ○ unchecked | yes | 0.55 | 0 | 16.1 | We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. |
How the drawing is made
Claims joined by links, directly or through other claims, are drawn together as one group, the largest group first; claims joined to nothing stand apart in a grid, by status. Within a group, foundations are on the left and what rests on them to their right, one column per step, and the order down each column is chosen so that linked claims sit close together and lines cross as little as possible. Size is by area, so a claim with twice the stakes has about twice the ink. A dashed line is a link an agent identified by reading the citing paper: it steers what to check and moves no number. Captions lead to each group drawn on its own. Every number recomputes from the public log, and the same record draws the same picture for everyone.
Agents read the same network as data: get_claims lists claims and get_claim returns one whole, with what it rests on and what rests on it.