Stochastic Flows and Geometric Optimization on the Orthogonal Group
Authors: Krzysztof Choromanski, David Cheikhi, Jared Davis, Valerii Likhosherstov, Achille Nazaret, Achraf Bahamou, Xingyou Song, Mrugank Akarte, Jack Parker-Holder, Jacob Bergquist, Yuan Gao, Aldo Pacchiano, Tamas Sarlos, Adrian Weller, Vikas Sindhwani
ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We theoretically and experimentally demonstrate that our methods can be applied in various fields of machine learning including deep, convolutional and recurrent neural networks, reinforcement learning, normalizing flows and metric learning. We demonstrate broad applicability of our methods by showing strong performance on the seemingly unrelated tasks of learning world models to obtain stable policies for the most difficult Humanoid agent from Open AI Gym and improving convolutional neural networks. |
| Researcher Affiliation | Collaboration | 1Google Brain Robotics, New York, USA 2Department of Industrial Engineering and Operations Research, Columbia University, New York, USA 3Department of Computer Science, Columbia University, New York, USA 4Department of Engineering, University of Cambridge, Cambridge, United Kingdom 5Department of Engineering, University of Oxford, Oxford, United Kingdom 6Department of Computer Science, University of California, Berkeley, USA 7Google Research, Mountain View, USA 8The Alan Turing Institute, London, United Kingdom. Correspondence to: Krzysztof Choromanski <kchoro@google.com>. |
| Pseudocode | Yes | Algorithm 1 Constructing tournament T Ph(T ) and ΩT |
| Open Source Code | No | The paper does not provide a specific repository link, explicit code release statement, or indicate that code is included in supplementary materials for the methodology described. |
| Open Datasets | Yes | We demonstrate the effectiveness of our approach in optimizing policies for a variety of continuous RL tasks from the Open AI Gym (Humanoid, Walker2d, Half Cheetah) and DM Control Suite (Reacher : Hard, Hopper Stand and Swimmer : 15). For MNIST, we used a 2-layer MLP... For CIFAR10, we used a Plain Net-110... |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning for training, validation, and test sets. While it mentions 'Training/Test accuracy' for MNIST, it does not specify the splits. |
| Hardware Specification | No | The paper mentions 'For all environments distributed optimization is conducted on 800 machines.' but does not provide specific hardware details such as GPU/CPU models, processor types, or memory amounts. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment. |
| Experiment Setup | Yes | Setting: For each method we run optimization for s = 10 random seeds (resulting in 240 experiments) and hidden state of sizes h = 200, 400 (similar results in both settings). For all environments distributed optimization is conducted on 800 machines. For MNIST, we used a 2-layer MLP with each layer of width 100, and tanh activations... For CIFAR10, we used a Plain Net-110 (Res Net-110 (He et al., 2016) without residual layers)... |