Cognitive Deficit of Deep Learning in Numerosity
Authors: Xiaolin Wu, Xi Zhang, Xiao Shu1303-1310
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | extensive experiments of the type of cognitive psychology demonstrate that the examples-driven black box DL cannot see through superficial variations in visual representations and distill the abstract notion of natural number, a task that children perform with high accuracy and confidence. We conduct a family of cognitive experiments to test if DL can, under various levels of supervision, learn the simple concept of natural numbers by observing sample images containing a varying number of objects in different positions, orientations, sizes, shapes and colors. |
| Researcher Affiliation | Academia | Department of Electronic Engineering, Shanghai Jiao Tong University, China Department of Electrical & Computer Engineering, Mc Master University, Canada |
| Pseudocode | No | No structured pseudocode or explicitly labeled algorithm blocks were found. The paper describes the proposed deterministic DCNN algorithm conceptually and visually through figures and text. |
| Open Source Code | No | The paper does not provide a specific link or explicit statement about the release of its source code. |
| Open Datasets | No | The paper describes synthetic training images (e.g., 'The training images for class n (n = 1, 2, , 6) consist of n white solid circles in black background (see Fig. 1)'). However, it does not provide concrete access information (link, DOI, specific citation with authors/year) for a publicly available or open dataset. |
| Dataset Splits | No | The paper mentions 'test images' but does not explicitly provide specific train/validation/test dataset split information (e.g., percentages, sample counts, or citations to predefined splits) needed to reproduce the data partitioning. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers (e.g., specific libraries like PyTorch or TensorFlow, or their versions). |
| Experiment Setup | No | The paper describes the DCNN configuration (Table 1) but does not provide specific experimental setup details such as concrete hyperparameter values (e.g., learning rate, batch size, number of epochs) or specific optimizer settings. |