An implicit function learning approach for parametric modal regression

Authors: Yangchen Pan, Ehsan Imani, Amir-massoud Farahmand, Martha White

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically demonstrate on several synthetic problems that our method (i) can learn multi-valued functions and produce the conditional modes, (ii) scales well to high-dimensional inputs, and (iii) can even be more effective for certain uni-modal problems, particularly for high-frequency functions. We demonstrate that our method is competitive in a real-world modal regression problem and two regular regression datasets.
Researcher Affiliation Academia Yangchen Pan, Ehsan Imani Univ. of Alberta & Amii {pan6,imani}@ualberta.ca; Amir-massoud Farahmand Vector Institute & Univ. of Toronto farahmand@vectorinstitute.ai; Martha White Univ. of Alberta whitem@ualberta.ca
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement or link regarding the availability of open-source code for its methodology.
Open Datasets Yes We construct a modal regression dataset from the Medical Cost Personal Dataset (Lantz, 2013) using the following steps.
Dataset Splits No The paper states "All of our results are averaged over 5 runs and for each run, the data is randomly split into training and testing sets." but does not explicitly mention or quantify a validation set split.
Hardware Specification No The paper does not provide specific details about the hardware specifications (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions "Tensor Flow" as a software reference but does not specify any software dependencies with version numbers that are directly used for its experiments.
Experiment Setup Yes For both our algorithm (Implicit) and MDN, we use a 16 16 tanh units NN and train by stochastic gradient descent with mini-batch size 128. We optimize both algorithms by sweeping the learning rate from {0.01, 0.001, 0.0001}.