Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Video Pixel Networks
Authors: Nal Kalchbrenner, Aäron Oord, Karen Simonyan, Ivo Danihelka, Oriol Vinyals, Alex Graves, Koray Kavukcuoglu
ICML 2017 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate VPNs on two benchmarks. The first is the Moving MNIST dataset... The second benchmark is the Robotic Pushing dataset... |
| Researcher Affiliation | Industry | 1Google Deep Mind, London, UK. |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating that the source code for the methodology is open-source or publicly available. |
| Open Datasets | Yes | The first is the Moving MNIST dataset (Srivastava et al., 2015a)... The second benchmark is the Robotic Pushing dataset (Finn et al., 2016) |
| Dataset Splits | Yes | The Robotic Pushing dataset consists of sequences of 20 frames... The data consists of a training set of 50000 sequences, a validation set, and two test sets of 1500 sequences each |
| Hardware Specification | No | The paper does not specify any particular hardware used for running the experiments (e.g., GPU models, CPU types, or cloud compute instances). |
| Software Dependencies | No | The paper mentions 'RMSProp for the optimization' but does not specify software versions for any libraries, frameworks, or programming languages used. |
| Experiment Setup | Yes | We train the models for 300000 steps with 20-frame sequences predicting the last 10 frames of each sequence. Each step corresponds to a batch of 64 sequences. We use RMSProp for the optimization with an initial learning rate of 3 x 10−4 and multiply the learning rate by 0.3 when learning flatlines. |