AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

The Tunnel Effect: Building Data Representations in Deep Neural Networks

Wojciech Masarczyk, Mateusz Ostaszewski, Ehsan Imani, Razvan Pascanu, Piotr Miłoś, Tomasz Trzcinski

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use deeper layers after the representation extractor as a compressing tunnelOverparameterized image classifiers · CIFAR-10, CIFAR-100, and CINIC-10linear-probe accuracy across layersNo clear effect
Continue through the tunnel rather than using features at its entryVGG-19 and ResNet-34 trained on CIFAR-10 · OOD datasetslinear-probe accuracyHurt
Increase the number of classes in the source taskVGG-19 and ResNet-34 · CIFAR-100 subsets; OOD linear probingOOD accuracyImproved
Increase network depth while holding the task fixed (ablation)MLP · CIFAR-10extractor lengthNo clear effect
Increase network width (ablation)VGG networks · CIFAR-10tunnel lengthImproved
Increase network width from 30% to 100% (ablation)ResNet-34 · CIFAR-10layer at 95% of final probing accuracy: 19 { ($56\%)} -> 21 { ($61\%)}Mixed
Train on a dataset with fewer classes (ablation)VGG-19 and ResNet-34 · CIFAR-100 class subsetstunnel lengthImproved
Train on CINIC-10 instead of CIFAR-10, with more training samples but the same number of classes (ablation)VGG-19 and ResNet-34 · CINIC-10 versus CIFAR-10tunnel lengthNo clear effect
Reset tunnel weights to their pre-optimization values (ablation)VGG-19 · CIFAR-10model performanceHurt
Replace the tunnel from one task with the tunnel from another task (ablation)VGG-19 task-specific extractor-tunnel combinations · CIFAR-10, two five-class tasksaccuracy: 92.04% -> 92.5%; 56.8% -> 58.04 %No clear effect
Attach a linear probe to the extractor instead of the extractor plus tunnel (ablation)VGG-19 after Task 2 · CIFAR-10, Task 1accuracy: 56.1% -> 74.4%Improved
Train a shorter VGG network from scratch (ablation)VGG-19 · CIFAR-10, sequential two-task learningcatastrophic forgettingImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy