The Teaching Dimension of Linear Learners
Authors: Ji Liu, Xiaojin Zhu, Hrag Ohannessian
ICML 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We illustrate some of the teaching dimensions by examples. These numerical experiments complement the theory to help build intuition and understanding. 6.1 Sometimes SVM can be Taught by One Item... We train our SVM with training set D0 by solving the standard SVM optimization problem... This was implemented with CVX (Grant & Boyd, 2014; 2008). ... We implement this teaching-set-finding problem in GAMS (using the nonlinear solvers SNOPT and MINOS)... |
| Researcher Affiliation | Academia | Ji Liu JI.LIU.UWISC@GMAIL.COM University of Rochester, Rochester, NY 14627 USA Xiaojin Zhu JERRYZHU@CS.WISC.EDU H. Gorune Ohannessian OHANNESSIAN@WISC.EDU University of Wisconsin-Madison, Madison, WI 53706 USA |
| Pseudocode | No | The paper provides mathematical formulations and propositions but does not include any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions using CVX and GAMS, which are third-party software packages, but it does not state that the authors' own implementation code is open-source or provide a link to it. |
| Open Datasets | No | The paper constructs synthetic teaching sets for its numerical examples, such as D0 = {x1 = [5e-5 ... 5e-5], y1 = 1} or specific x+ and x- points. It does not use or provide access to a publicly available or open dataset for training. |
| Dataset Splits | No | The paper does not specify training, validation, or test dataset splits. It constructs specific teaching sets to demonstrate its theoretical findings. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments. It only mentions using CVX and GAMS for implementation. |
| Software Dependencies | Yes | This was implemented with CVX (Grant & Boyd, 2014; 2008). ... We implement this teaching-set-finding problem in GAMS (using the nonlinear solvers SNOPT and MINOS). The bibliography lists: GAMS. General Algebraic Modeling System (GAMS) Release 24.2.1. Washington, DC, USA, 2013. |
| Experiment Setup | Yes | We consider a homogeneous SVM in Rd with regularization weight λ = 5e-5. ... The learner is an inhomogeneous SVM (20) with λ = 1. ... a different SVM learner with λ = 3. |