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; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]
Blending Learning and Inference in Conditional Random Fields
Authors: Tamir Hazan, Alexander G. Schwing, Raquel Urtasun
JMLR 2016 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The proposed algorithm currently achieves the state-of-the-art in various computer vision applications. In Section 6, titled 'Experiments', the authors demonstrate the effectiveness of their approach on a 3D scene understanding application, evaluating it on the layout dataset (Hedau et al., 2009) and presenting results like 'Test set percentage pixel error' and 'Test set pixel classification error' in figures. |
| Researcher Affiliation | Academia | Tamir Hazan EMAIL Technion Israel Institute of Technology Technion City, Haifa, 3200, Israel; Alexander G. Schwing EMAIL Electrical and Computer Engineering and Coordinated Science Laboratory, University of Illinois at Urbana-Champaign Urbana, IL 61801; Raquel Urtasun EMAIL University of Toronto 40 St. George Street, Toronto, ON, M5S 2E4. All listed affiliations are academic institutions. |
| Pseudocode | Yes | The paper includes 'Algorithm 1 Sum-product belief propagation', 'Algorithm 2 Norm-Product Belief Propagation', and a block labeled 'Blending learning and inference' in Figure 2, which outlines the steps of the algorithm. |
| Open Source Code | Yes | The code is publicly available on http://www.alexander-schwing.de/ projects General Structured Prediction Latent Variables.php. |
| Open Datasets | Yes | The paper states: 'For this purpose we elaborate more carefully on a 3D scene understanding application (Schwing et al., 2012a) evaluated on the well known layout dataset (Hedau et al., 2009) containing 314 indoor images.' and refers to the 'bedroom data set of Hedau et al. (2010)'. |
| Dataset Splits | Yes | The paper states: 'Due to the involved randomness it failed in our case on 9 training images and was successful on all 105 test set instances.' This explicitly mentions training and test sets and provides a count for the test set. |
| Hardware Specification | No | The paper discusses performance in terms of 'wall clock time' (Figure 6) but does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions an 'efficient C++ implementation with a Matlab wrapper' but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | The paper provides details on the experimental setup, including: 'The step size η is set to guarantee convergence (e.g., corresponding to the Lipschitz constant or the Armijo rule.)'; 'In our experiments we learn and infer with the same counting numbers.'; 'standard learning approach (Standard 20), which performs 20 message passing iterations'. |