Deep Layers as Stochastic Solvers
Authors: Adel Bibi, Bernard Ghanem, Vladlen Koltun, Rene Ranftl
ICLR 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experiments with standard convolutional networks applied to the CIFAR-10 and CIFAR-100 datasets and show that replacing a block of layers with multiple iterations of the corresponding solver, with step size set via L, consistently improves classification accuracy. |
| Researcher Affiliation | Collaboration | Adel Bibi KAUST Bernard Ghanem KAUST Vladlen Koltun Intel Labs Ren e Ranftl Intel Labs |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code. |
| Open Datasets | Yes | We conduct experiments on CIFAR-10 and CIFAR-100 (Krizhevsky & Hinton, 2009). |
| Dataset Splits | Yes | In all experiments, training was conducted on 90% of the training set while 10% was left for validation. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | Yes | We used stochastic gradient descent with a momentum of 0.9 and a weight decay of 5 10 4. The learning rate was set to (10 2, 10 3, 10 4) for the first, second, and third 100 epochs, respectively. For finetuning, the learning rate was initially set to 10 3 and reduced to 10 4 after 100 epochs... Finally, to guarantee convergence of the stochastic solvers, we add the strongly convex function λ 2 X 2 F to the finite sum in (5), where we set λ = 10 3 in all experiments. |