More Like this

1. Identifying Bias in Deep Neural Networks Using Image Transforms

   https://doi.org/10.3390/computers13120341  

   Erukude, Sai Teja; Joshi, Akhil; Shamir, Lior (December 2024, Computers)

   CNNs have become one of the most commonly used computational tools in the past two decades. One of the primary downsides of CNNs is that they work as a “black box”, where the user cannot necessarily know how the image data are analyzed, and therefore needs to rely on empirical evaluation to test the efficacy of a trained CNN. This can lead to hidden biases that affect the performance evaluation of neural networks, but are difficult to identify. Here we discuss examples of such hidden biases in common and widely used benchmark datasets, and propose techniques for identifying dataset biases that can affect the standard performance evaluation metrics. One effective approach to identify dataset bias is to perform image classification by using merely blank background parts of the original images. However, in some situations, a blank background in the images is not available, making it more difficult to separate foreground or contextual information from the bias. To overcome this, we propose a method to identify dataset bias without the need to crop background information from the images. The method is based on applying several image transforms to the original images, including Fourier transform, wavelet transforms, median filter, and their combinations. These transforms are applied to recover background bias information that CNNs use to classify images. These transformations affect the contextual visual information in a different manner than it affects the systemic background bias. Therefore, the method can distinguish between contextual information and the bias, and can reveal the presence of background bias even without the need to separate sub-image parts from the blank background of the original images. The code used in the experiments is publicly available. 

   more » « less
2. Contextually Guided Convolutional Neural Networks for Learning Most Transferable Representations

   https://doi.org/10.1109/ISM55400.2022.00047  

   Kursun, Olcay; Dinc, Semih; Favorov, Oleg V. (December 2022, 2022 IEEE International Symposium on Multimedia (ISM))

   Implementing local contextual guidance principles in a single-layer CNN architecture, we propose an efficient algorithm for developing broad-purpose representations (i.e., representations transferable to new tasks without additional training) in shallow CNNs trained on limited-size datasets. A contextually guided CNN (CG-CNN) is trained on groups of neighboring image patches picked at random image locations in the dataset. Such neighboring patches are likely to have a common context and therefore are treated for the purposes of training as belonging to the same class. Across multiple iterations of such training on different context-sharing groups of image patches, CNN features that are optimized in one iteration are then transferred to the next iteration for further optimization, etc. In this process, CNN features acquire higher pluripotency, or inferential utility for any arbitrary classification task. In our applications to natural images and hyperspectral images, we find that CG-CNN can learn transferable features similar to those learned by the first layers of the well-known deep networks and produce favorable classification accuracies. 

   more » « less
3. An automated approach for improving the inference latency and energy efficiency of pretrained CNNs by removing irrelevant pixels with focused convolutions

   https://doi.org/10.1109/ASP-DAC58780.2024.10473884  

   Tung, Caleb; Eliopoulos, Nicholas; Jajal, Purvish; Ramshankar, Gowri; Yang, Cheng-Yun; Synovic, Nicholas; Zhang, Xuecen; Chaudhary, Vipin; Thiruvathukal, George K; Lu, Yung-Hsiang (January 2024, Asia and South Pacific Design Automation Conference (ASP-DAC))

   Computer vision often uses highly accurate Convolutional Neural Networks (CNNs), but these deep learning models are associated with ever-increasing energy and computation requirements. Producing more energy-efficient CNNs often requires model training which can be cost-prohibitive. We propose a novel, automated method to make a pretrained CNN more energyefficient without re-training. Given a pretrained CNN, we insert a threshold layer that filters activations from the preceding layers to identify regions of the image that are irrelevant, i.e. can be ignored by the following layers while maintaining accuracy. Our modified focused convolution operation saves inference latency (by up to 25%) and energy costs (by up to 22%) on various popular pretrained CNNs, with little to no loss in accuracy 

   more » « less
4. On Human-like Biases in Convolutional Neural Networks for the Perception of Slant from Texture

   https://doi.org/10.1145/3613451  

   Wang, Yuanhao; Zhang, Qian; Aubuchon, Celine; Kemp, Jovan; Domini, Fulvio; Tompkin, James (August 2023, ACM Transactions on Applied Perception)

   Depth estimation is fundamental to 3D perception, and humans are known to have biased estimates of depth. This study investigates whether convolutional neural networks (CNNs) can be biased when predicting the sign of curvature and depth of surfaces of textured surfaces under different viewing conditions (field of view) and surface parameters (slant and texture irregularity). This hypothesis is drawn from the idea that texture gradients described by local neighborhoods—a cue identified in human vision literature—are also representable within convolutional neural networks. To this end, we trained both unsupervised and supervised CNN models on the renderings of slanted surfaces with random Polka dot patterns and analyzed their internal latent representations. The results show that the unsupervised models have similar prediction biases as humans across all experiments, while supervised CNN models do not exhibit similar biases. The latent spaces of the unsupervised models can be linearly separated into axes representing field of view and optical slant. For supervised models, this ability varies substantially with model architecture and the kind of supervision (continuous slant vs. sign of slant). Even though this study says nothing of any shared mechanism, these findings suggest that unsupervised CNN models can share similar predictions to the human visual system. Code: github.com/brownvc/Slant-CNN-Biases 

   more » « less
5. An Information-theoretic Visual Analysis Framework for Convolutional Neural Networks

   Jingyi Shen, Han-Wei Shen (November 2021, Smart Tools and Apps for Graphics (STAG))

   Despite the great success of Convolutional Neural Networks (CNNs) in Computer Vision and Natural Language Processing, the working mechanism behind CNNs is still under extensive discussion and research. Driven by strong demand for the theoretical explanation of neural networks, some researchers utilize information theory to provide insight into the black-box model. However, to the best of our knowledge, employing information theory to quantitatively analyze and qualitatively visualize neural networks has not been extensively studied in the visualization community. In this paper, we combine information entropies and visualization techniques to shed light on how CNN works. Specifically, we first introduce a data model to organize the data that can be extracted from CNN models. Then we propose two ways to calculate entropy under different circumstances. To provide a fundamental understanding of the basic building blocks of CNNs (e.g., convolutional layers, pooling layers, normalization layers) from an information-theoretic perspective, we develop a visual analysis system, CNNSlicer. CNNSlicer allows users to interactively explore the amount of information changes inside the model. With case studies on the widely used benchmark datasets (MNIST and CIFAR-10), we demonstrate the effectiveness of our system in opening the black-box of CNNs. 

   more » « less