Tensor-Train Recurrent Neural Networks for Video Classification
Authors: Yinchong Yang, Denis Krompass, Volker Tresp
ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We test our model on classification tasks using multiple real-world video datasets and achieve competitive performances with state-of-the-art models, even though our model architecture is orders of magnitude less complex.In the following, we present our experiments conducted on three large video datasets. These empirical results demonstrate that the integration of the Tensor-Train Layer in plain RNN architectures such as a tensorized LSTM or GRU boosts the classification quality of these models tremendously when directly exposed to high-dimensional input data, such as video data. |
| Researcher Affiliation | Collaboration | 1Ludwig Maximilian University of Munich, Germany 2Siemens AG, Corporate Technology, Germany. |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The source codes of our TT-RNN implementations and all the experiments in Sec. 4 are publicly available at https://github.com/Tuyki/TT_RNN. |
| Open Datasets | Yes | UCF11 Data (Liu et al., 2009), Hollywood2 Data (Marszałek et al., 2009), Youtube Celebrities Face Data (Kim et al., 2008). |
| Dataset Splits | Yes | We follow (Liu et al., 2013) and perform for each experimental setting a 5-fold cross validation with mutual exclusive data splits. |
| Hardware Specification | Yes | The models were trained on a Quad core Intel R Xeon R E74850 v2 2.30GHz Processor to a maximum of 100 epochs.The models were trained on an NVIDIA Tesla K40c Processor to a maximum of 500 epochs. |
| Software Dependencies | No | The paper mentions 'Theano (Bastien et al., 2012) and deployed in Keras (Chollet, 2015)' but does not provide specific version numbers for these software dependencies. |
| Experiment Setup | Yes | We applied 0.25 Dropout (Srivastava et al., 2014) for both input-to-hidden and hidden-to-hidden mappings in plain GRU and LSTM as well as their respective TT modifications; and 0.01 ridge regularization for the single-layered classifier. We used the Adam (Kingma & Ba, 2014) step rule for the updates with an initial learning rate 0.001. |