From fully connected neural networks to convolutional neural networks, the learned parameters within a neural network have been primarily relegated to the linear parameters (e.g., convolutional filters). The non-linear functions (e.g., activation functions) have largely remained, with few exceptions in recent years, parameter-less, static throughout training, and seen limited variation in design. Largely ignored by the deep learning community, radial basis function (RBF) networks provide an interesting mechanism for learning more complex non-linear activation functions in addition to the linear parameters in a network. However, the interest in RBF networks has waned over time due to the difficulty of integrating RBFs into more complex deep neural network architectures in a tractable and stable manner. In this work, we present a novel approach that enables end-to-end learning of deep RBF networks with fully learnable activation basis functions in an automatic and tractable manner. We demonstrate that our approach for enabling the use of learnable activation basis functions in deep neural networks, which we will refer to as DeepLABNet, is an effective tool for automated activation function learning within complex network architectures.
There has been a significant surge of interest recently around the concept of explainable artificial intelligence (XAI), where the goal is to produce an interpretation for a decision made by a machine learning algorithm. Of particular interest is the interpretation of how deep neural networks make decisions, given the complexity and `black box' nature of such networks. Given the infancy of the field, there has been very limited exploration into the assessment of the performance of explainability methods, with most evaluations centered around subjective visual interpretation of the produced interpretations. In this study, we explore a more machine-centric strategy for quantifying the performance of explainability methods on deep neural networks via the notion of decision-making impact analysis. We introduce two quantitative performance metrics: i) Impact Score, which assesses the percentage of critical factors with either strong confidence reduction impact or decision changing impact, and ii) Impact Coverage, which assesses the percentage coverage of adversarially impacted factors in the input. A comprehensive analysis using this approach was conducted on several state-of-the-art explainability methods (LIME, SHAP, Expected Gradients, GSInquire) on a ResNet-50 deep convolutional neural network using a subset of ImageNet for the task of image classification. Experimental results show that the critical regions identified by LIME within the tested images had the lowest impact on the decision-making process of the network (~38%), with progressive increase in decision-making impact for SHAP (~44%), Expected Gradients (~51%), and GSInquire (~76%). While by no means perfect, the hope is that the proposed machine-centric strategy helps push the conversation forward towards better metrics for evaluating explainability methods and improve trust in deep neural networks.
Malnutrition impacts quality of life and places annually-recurring burden on the health care system. Half of older adults are at risk for malnutrition in long-term care (LTC). Monitoring and measuring nutritional intake is paramount yet involves time-consuming and subjective visual assessment, limiting current methods' reliability. The opportunity for automatic image-based estimation exists. Some progress outside LTC has been made (e.g., calories consumed, food classification), however, these methods have not been implemented in LTC, potentially due to a lack of ability to independently evaluate automatic segmentation methods within the intake estimation pipeline. Here, we propose and evaluate a novel fully-automatic semantic segmentation method for pixel-level classification of food on a plate using a deep convolutional neural network (DCNN). The macroarchitecture of the DCNN is a multi-scale encoder-decoder food network (EDFN) architecture comprising a residual encoder microarchitecture, a pyramid scene parsing decoder microarchitecture, and a specialized per-pixel food/no-food classification layer. The network was trained and validated on the pre-labelled UNIMIB 2016 food dataset (1027 tray images, 73 categories), and tested on our novel LTC plate dataset (390 plate images, 9 categories). Our fully-automatic segmentation method attained similar intersection over union to the semi-automatic graph cuts (91.2% vs. 93.7%). Advantages of our proposed system include: testing on a novel dataset, decoupled error analysis, no user-initiated annotations, with similar segmentation accuracy and enhanced reliability in terms of types of segmentation errors. This may address several short-comings currently limiting utility of automated food intake tracking in time-constrained LTC and hospital settings.
There has been a significant surge of interest recently around the concept of explainable artificial intelligence (XAI), where the goal is to produce an interpretation for a decision made by a machine learning algorithm. Of particular interest is the interpretation of how deep neural networks make decisions, given the complexity and `black box' nature of such networks. Given the infancy of the field, there has been very limited exploration into the assessment of the performance of explainability methods, with most evaluations centered around subjective visual interpretation of the produced interpretations. In this study, we explore a more machine-centric strategy for quantifying the performance of explainability methods on deep neural networks via the notion of decision-making impact analysis. We introduce two quantitative performance metrics: i) Impact Score, which assesses the percentage of critical factors with either strong confidence reduction impact or decision changing impact, and ii) Impact Coverage, which assesses the percentage coverage of adversarially impacted factors in the input. A comprehensive analysis using this approach was conducted on several state-of-the-art explainability methods (LIME, SHAP, Expected Gradients, GSInquire) on a ResNet-50 deep convolutional neural network using a subset of ImageNet for the task of image classification. Experimental results show that the critical regions identified by LIME within the tested images had the lowest impact on the decision-making process of the network (~38%), with progressive increase in decision-making impact for SHAP (~44%), Expected Gradients (~51%), and GSInquire (~76%). While by no means perfect, the hope is that the proposed machine-centric strategy helps push the conversation forward towards better metrics for evaluating explainability methods and improve trust in deep neural networks.
The design of compact deep neural networks is a crucial task to enable widespread adoption of deep neural networks in the real-world, particularly for edge and mobile scenarios. Due to the time-consuming and challenging nature of manually designing compact deep neural networks, there has been significant recent research interest into algorithms that automatically search for compact network architectures. A particularly interesting class of compact architecture search algorithms are those that are guided by baseline network architectures. Such algorithms have been shown to be significantly more computationally efficient than unguided methods. In this study, we explore the current state of compact architecture search for deep neural networks through both theoretical and empirical analysis of four different state-of-the-art compact architecture search algorithms: i) group lasso regularization, ii) variational dropout, iii) MorphNet, and iv) Generative Synthesis. We examine these methods in detail based on a number of different factors such as efficiency, effectiveness, and scalability. Furthermore, empirical evaluations are conducted to compare the efficacy of these compact architecture search algorithms across three well-known benchmark datasets. While by no means an exhaustive exploration, we hope that this study helps provide insights into the interesting state of this relatively new area of research in terms of diversity and real, tangible gains already achieved in architecture design improvements. Furthermore, the hope is that this study would help in pushing the conversation forward towards a deeper theoretical and empirical understanding where the research community currently stands in the landscape of compact architecture search for deep neural networks, and the practical challenges and considerations in leveraging such approaches for operational usage.
Object detection remains an active area of research in the field of computer vision, and considerable advances and successes has been achieved in this area through the design of deep convolutional neural networks for tackling object detection. Despite these successes, one of the biggest challenges to widespread deployment of such object detection networks on edge and mobile scenarios is the high computational and memory requirements. As such, there has been growing research interest in the design of efficient deep neural network architectures catered for edge and mobile usage. In this study, we introduce YOLO Nano, a highly compact deep convolutional neural network for the task of object detection. A human-machine collaborative design strategy is leveraged to create YOLO Nano, where principled network design prototyping, based on design principles from the YOLO family of single-shot object detection network architectures, is coupled with machine-driven design exploration to create a compact network with highly customized module-level macroarchitecture and microarchitecture designs tailored for the task of embedded object detection. The proposed YOLO Nano possesses a model size of ~4.0MB (>15.1x and >8.3x smaller than Tiny YOLOv2 and Tiny YOLOv3, respectively) and requires 4.57B operations for inference (>34% and ~17% lower than Tiny YOLOv2 and Tiny YOLOv3, respectively) while still achieving an mAP of ~69.1% on the VOC 2007 dataset (~12% and ~10.7% higher than Tiny YOLOv2 and Tiny YOLOv3, respectively). Experiments on inference speed and power efficiency on a Jetson AGX Xavier embedded module at different power budgets further demonstrate the efficacy of YOLO Nano for embedded scenarios.
An effective deep learning development process is critical for widespread industrial adoption, particularly in the automotive sector. A typical industrial deep learning development cycle involves customizing and re-designing an off-the-shelf network architecture to meet the operational requirements of the target application, leading to considerable trial and error work by a machine learning practitioner. This approach greatly impedes development with a long turnaround time and the unsatisfactory quality of the created models. As a result, a development platform that can aid engineers in greatly accelerating the design and production of compact, optimized deep neural networks is highly desirable. In this joint industrial case study, we study the efficacy of the GenSynth AI-assisted AI design platform for accelerating the design of custom, optimized deep neural networks for autonomous driving through human-machine collaborative design. We perform a quantitative examination by evaluating 10 different compact deep neural networks produced by GenSynth for the purpose of object detection via a NASNet-based user network prototype design, targeted at a low-cost GPU-based accelerated embedded system. Furthermore, we quantitatively assess the talent hours and GPU processing hours used by the GenSynth process and three other approaches based on the typical industrial development process. In addition, we quantify the annual cloud cost savings for comprehensive testing using networks produced by GenSynth. Finally, we assess the usability and merits of the GenSynth process through user feedback. The findings of this case study showed that GenSynth is easy to use and can be effective at accelerating the design and production of compact, customized deep neural network.
Squeeze-and-excitation (SE) module enhances the representational power of convolution layers by adaptively re-calibrating channel-wise feature responses. However, the limitation of SE in terms of attention characterization lies in the loss of spatial information cues, making it less well suited for perception tasks with very high spatial inter-dependencies such as semantic segmentation. In this paper, we propose a novel squeeze-and-attention network (SANet) architecture that leverages a simple but effective squeeze-and-attention (SA) module to account for two distinctive characteristics of segmentation: i) pixel-group attention, and ii) pixel-wise prediction. Specifically, the proposed SA modules impose pixel-group attention on conventional convolution by introducing an 'attention' convolutional channel, thus taking into account spatial-channel inter-dependencies in an efficient manner. The final segmentation results are produced by merging outputs from four hierarchical stages of a SANet to integrate multi-scale contexts for obtaining enhanced pixel-wise prediction. Empirical experiments using two challenging public datasets validate the effectiveness of the proposed SANets, which achieved 83.2% mIoU (without COCO pre-training) on PASCAL VOC and a state-of-the-art mIoU of 54.4% on PASCAL Context.
Person re-identification (ReID) remains a very difficult challenge in computer vision, and critical for large-scale video surveillance scenarios where an individual could appear in different camera views at different times. There has been recent interest in tackling this challenge using cross-domain approaches, which leverages data from source domains that are different than the target domain. Such approaches are more practical for real-world widespread deployment given that they don't require on-site training (as with unsupervised or domain transfer approaches) or on-site manual annotation and training (as with supervised approaches). In this study, we take a systematic approach to establishing a large baseline source domain and target domain for cross-domain person ReID. We accomplish this by conducting a comprehensive analysis to study the similarities between source domains proposed in literature, and studying the effects of incrementally increasing the size of the source domain. This allows us to establish a balanced source domain and target domain split that promotes variety in both source and target domains. Furthermore, using lessons learned from the state-of-the-art supervised person re-identification methods, we establish a strong baseline method for cross-domain person ReID. Experiments show that a source domain composed of two of the largest person ReID domains (SYSU and MSMT) performs well across six commonly-used target domains. Furthermore, we show that, surprisingly, two of the recent commonly-used domains (PRID and GRID) have too few query images to provide meaningful insights. As such, based on our findings, we propose the following balanced baseline for cross-domain person ReID consisting of: i) a fixed multi-source domain consisting of SYSU, MSMT, Airport and 3DPeS, and ii) a multi-target domain consisting of Market-1501, DukeMTMC-reID, CUHK03, PRID, GRID and VIPeR.
The ImageNet dataset ushered in a flood of academic and industry interest in deep learning for computer vision applications. Despite its significant impact, there has not been a comprehensive investigation into the demographic attributes of images contained within the dataset. Such a study could lead to new insights on inherent biases within ImageNet, particularly important given it is frequently used to pretrain models for a wide variety of computer vision tasks. In this work, we introduce a model-driven framework for the automatic annotation of apparent age and gender attributes in large-scale image datasets. Using this framework, we conduct the first demographic audit of the 2012 ImageNet Large Scale Visual Recognition Challenge (ILSVRC) subset of ImageNet and the "person" hierarchical category of ImageNet. We find that 41.62% of faces in ILSVRC appear as female, 1.71% appear as individuals above the age of 60, and males aged 15 to 29 account for the largest subgroup with 27.11%. We note that the presented model-driven framework is not fair for all intersectional groups, so annotation are subject to bias. We present this work as the starting point for future development of unbiased annotation models and for the study of downstream effects of imbalances in the demographics of ImageNet. Code and annotations are available at: http://bit.ly/ImageNetDemoAudit