Fully supervised change detection methods require difficult to procure pixel-level labels, while weakly supervised approaches can be trained with image-level labels. However, most of these approaches require a combination of changed and unchanged image pairs for training. Thus, these methods can not directly be used for datasets where only changed image pairs are available. We present W-CDNet, a novel weakly supervised change detection network that can be trained with image-level semantic labels. Additionally, W-CDNet can be trained with two different types of datasets, either containing changed image pairs only or a mixture of changed and unchanged image pairs. Since we use image-level semantic labels for training, we simultaneously create a change mask and label the changed object for single-label images. W-CDNet employs a W-shaped siamese U-net to extract feature maps from an image pair which then get compared in order to create a raw change mask. The core part of our model, the Change Segmentation and Classification (CSC) module, learns an accurate change mask at a hidden layer by using a custom Remapping Block and then segmenting the current input image with the change mask. The segmented image is used to predict the image-level semantic label. The correct label can only be predicted if the change mask actually marks relevant change. This forces the model to learn an accurate change mask. We demonstrate the segmentation and classification performance of our approach and achieve top results on AICD and HRSCD, two public aerial imaging change detection datasets as well as on a Food Waste change detection dataset. Our code is available at https://github.com/PhiAbs/W-CDNet .
The success of deep learning methods led to significant breakthroughs in 3-D point cloud processing tasks with applications in remote sensing. Existing methods utilize convolutions that have some limitations, as they assume a uniform input distribution and cannot learn long-range dependencies. Recent works have shown that adding attention in conjunction with these methods improves performance. This raises a question: can attention layers completely replace convolutions? This paper proposes a fully attentional model - {\em Point Transformer}, for deriving a rich point cloud representation. The model's shape classification and retrieval performance are evaluated on a large-scale urban dataset - RoofN3D and a standard benchmark dataset ModelNet40. Extensive experiments are conducted to test the model's robustness to unseen point corruptions for analyzing its effectiveness on real datasets. The proposed method outperforms other state-of-the-art models in the RoofN3D dataset, gives competitive results in the ModelNet40 benchmark, and showcases high robustness to various unseen point corruptions. Furthermore, the model is highly memory and space efficient when compared to other methods.
This article proposes a Universal Activation Function (UAF) that achieves near optimal performance in quantification, classification, and reinforcement learning (RL) problems. For any given problem, the optimization algorithms are able to evolve the UAF to a suitable activation function by tuning the UAF's parameters. For the CIFAR-10 classification and VGG-8, the UAF converges to the Mish like activation function, which has near optimal performance $F_{1} = 0.9017\pm0.0040$ when compared to other activation functions. For the quantification of simulated 9-gas mixtures in 30 dB signal-to-noise ratio (SNR) environments, the UAF converges to the identity function, which has near optimal root mean square error of $0.4888 \pm 0.0032$ $\mu M$. In the BipedalWalker-v2 RL dataset, the UAF achieves the 250 reward in $961 \pm 193$ epochs, which proves that the UAF converges in the lowest number of epochs. Furthermore, the UAF converges to a new activation function in the BipedalWalker-v2 RL dataset.
Supervised learning is based on the assumption that the ground truth in the training data is accurate. However, this may not be guaranteed in real-world settings. Inaccurate training data will result in some unexpected predictions. In image classification, incorrect labels may cause the classification model to be inaccurate as well. In this paper, I am going to apply unsupervised techniques to the training data before training the classification network. A convolutional autoencoder is applied to encode and reconstruct images. The encoder will project the image data on to latent space. In the latent space, image features are preserved in a lower dimension. The assumption is that data samples with similar features are likely to have the same label. Noised samples can be classified in the latent space by the Density-Base Scan (DBSCAN) clustering algorithm. These incorrectly labeled data are visualized as outliers in the latent space. Therefore, the outliers identified by the DBSCAN algorithm can be classified as incorrectly labeled samples. After the outliers are detected, all the outliers are treated as mislabeled data samples and removed from the dataset. Thus the training data can be directly used in training the supervised learning network. The algorithm can detect and remove above 67\% of mislabeled data in the experimental dataset.
In this paper, we propose a study on multi-modal (audio and video) action spotting and classification in soccer videos. Action spotting and classification are the tasks that consist in finding the temporal anchors of events in a video and determine which event they are. This is an important application of general activity understanding. Here, we propose an experimental study on combining audio and video information at different stages of deep neural network architectures. We used the SoccerNet benchmark dataset, which contains annotated events for 500 soccer game videos from the Big Five European leagues. Through this work, we evaluated several ways to integrate audio stream into video-only-based architectures. We observed an average absolute improvement of the mean Average Precision (mAP) metric of $7.43\%$ for the action classification task and of $4.19\%$ for the action spotting task.
We propose a new character-based text classification framework for non-alphabetic languages, such as Chinese and Japanese. Our framework consists of a variational character encoder (VCE) and character-level text classifier. The VCE is composed of a $\beta$-variational auto-encoder ($\beta$-VAE) that learns the proposed glyph-aware disentangled character embedding (GDCE). Since our GDCE provides zero-mean unit-variance character embeddings that are dimensionally independent, it is applicable for our interpretable data augmentation, namely, semantic sub-character augmentation (SSA). In this paper, we evaluated our framework using Japanese text classification tasks at the document- and sentence-level. We confirmed that our GDCE and SSA not only provided embedding interpretability but also improved the classification performance. Our proposal achieved a competitive result to the state-of-the-art model while also providing model interpretability. Our code is available on https://github.com/IyatomiLab/GDCE-SSA
Convolutional Neural Networks can be designed with different levels of complexity depending upon the task at hand. This paper analyzes the effect of dimensional changes to the CNN architecture on its performance on the task of Histopathological Cancer Classification. The research starts with a baseline 10-layer CNN model with (3 X 3) convolution filters. Thereafter, the baseline architecture is scaled in multiple dimensions including width, depth, resolution and a combination of all of these. Width scaling involves inculcating greater number of neurons per CNN layer, whereas depth scaling involves deepening the hierarchical layered structure. Resolution scaling is performed by increasing the dimensions of the input image, and compound scaling involves a hybrid combination of width, depth and resolution scaling. The results indicate that histopathological cancer scans are very complex in nature and hence require high resolution images fed to a large hierarchy of Convolution, MaxPooling, Dropout and Batch Normalization layers to extract all the intricacies and perform perfect classification. Since compound scaling the baseline model ensures that all the three dimensions: width, depth and resolution are scaled, the best performance is obtained with compound scaling. This research shows that better performance of CNN models is achieved by compound scaling of the baseline model for the task of Histopathological Cancer Classification.
Crop failure owing to pests & diseases are inherent within Indian agriculture, leading to annual losses of 15 to 25% of productivity, resulting in a huge economic loss. This research analyzes the performance of various optimizers for predictive analysis of plant diseases with deep learning approach. The research uses Convolutional Neural Networks for classification of farm or plant leaf samples of 3 crops into 15 classes. The various optimizers used in this research include RMSprop, Adam and AMSgrad. Optimizers Performance is visualised by plotting the Training and Validation Accuracy and Loss curves, ROC curves and Confusion Matrix. The best performance is achieved using Adam optimizer, with the maximum validation accuracy being 98%. This paper focuses on the research analysis proving that plant diseases can be predicted and pre-empted using deep learning methodology with the help of satellite, drone based or mobile based images that result in reducing crop failure and agricultural losses.
With the prevalence of Diabetes, the Diabetes Mellitus Retinopathy (DR) is becoming a major health problem across the world. The long-term medical complications arising due to DR have a significant impact on the patient as well as the society, as the disease mostly affects individuals in their most productive years. Early detection and treatment can help reduce the extent of damage to the patients. The rise of Convolutional Neural Networks for predictive analysis in the medical field paves the way for a robust solution to DR detection. This paper studies the performance of several highly efficient and scalable CNN architectures for Diabetic Retinopathy Classification with the help of Transfer Learning. The research focuses on VGG16, Resnet50 V2 and EfficientNet B0 models. The classification performance is analyzed using several performance metrics including True Positive Rate, False Positive Rate, Accuracy, etc. Also, several performance graphs are plotted for visualizing the architecture performance including Confusion Matrix, ROC Curve, etc. The results indicate that Transfer Learning with ImageNet weights using VGG 16 model demonstrates the best classification performance with the best Accuracy of 95%. It is closely followed by ResNet50 V2 architecture with the best Accuracy of 93%. This paper shows that predictive analysis of DR from retinal images is achieved with Transfer Learning on Convolutional Neural Networks.
Sparse approximation is the problem to find the sparsest linear combination for a signal from a redundant dictionary, which is widely applied in signal processing and compressed sensing. In this project, I manage to implement the Orthogonal Matching Pursuit (OMP) algorithm and Sparse Representation-based Classification (SRC) algorithm, then use them to finish the task of masked image classification with majority voting. Here the experiment was token on the AR data-set, and the result shows the superiority of OMP algorithm combined with SRC algorithm over masked face image classification with an accuracy of 98.4%.
Voxel-based 3D object classification has been thoroughly studied in recent years. Most previous methods convert the classic 2D convolution into a 3D form that will be further applied to objects with binary voxel representation for classification. However, the binary voxel representation is not very effective for 3D convolution in many cases. In this paper, we propose a hybrid cascade architecture for voxel-based 3D object classification. It consists of three stages composed of fully connected and convolutional layers, dealing with easy, moderate, and hard 3D models respectively. Both accuracy and speed can be balanced in our proposed method. By giving each voxel a signed distance value, an obvious gain regarding the accuracy can be observed. Besides, the mean inference time can be speeded up hugely compared with the state-of-the-art point cloud and voxel based methods.
Text classification has been one of the earliest problems in NLP. Over time the scope of application areas has broadened and the difficulty of dealing with new areas (e.g., noisy social media content) has increased. The problem-solving strategy switched from classical machine learning to deep learning algorithms. One of the recent deep neural network architecture is the Transformer. Models designed with this type of network and its variants recently showed their success in many downstream natural language processing tasks, especially for resource-rich languages, e.g., English. However, these models have not been explored fully for Bangla text classification tasks. In this work, we fine-tune multilingual transformer models for Bangla text classification tasks in different domains, including sentiment analysis, emotion detection, news categorization, and authorship attribution. We obtain the state of the art results on six benchmark datasets, improving upon the previous results by 5-29% accuracy across different tasks.
Deep learning methods achieve great success in many areas due to their powerful feature extraction capabilities and end-to-end training mechanism, and recently they are also introduced for radio signal modulation classification. In this paper, we propose a novel deep learning framework called SigNet, where a signal-to-matrix (S2M) operator is adopted to convert the original signal into a square matrix first and is co-trained with a follow-up CNN architecture for classification. This model is further accelerated by integrating 1D convolution operators, leading to the upgraded model SigNet2.0. The simulations on two signal datasets show that both SigNet and SigNet2.0 outperform a number of well-known baselines. More interestingly, our proposed models behave extremely well in small-sample learning when only a small training dataset is provided. They can achieve a relatively high accuracy even when 1\% training data are kept, while other baseline models may lose their effectiveness much more quickly as the datasets get smaller. Such result suggests that SigNet/SigNet2.0 could be extremely useful in the situations where labeled signal data are difficult to obtain. The visualization of the output features of our models demonstrates that our model can well divide different modulation types of signals in the feature hyper-space.
Automated cattle activity classification allows herders to continuously monitor the health and well-being of livestock, resulting in increased quality and quantity of beef and dairy products. In this paper, a sequential deep neural network is used to develop a behavioural model and to classify cattle behaviour and activities. The key focus of this paper is the exploration of a joint time-frequency domain representation of the sensor data, which is provided as the input to the neural network classifier. Our exploration is based on a real-world data set with over 3 million samples, collected from sensors with a tri-axial accelerometer, magnetometer and gyroscope, attached to collar tags of 10 dairy cows and collected over a one month period. The key results of this paper is that the joint time-frequency data representation, even when used in conjunction with a relatively basic neural network classifier, can outperform the best cattle activity classifiers reported in the literature. With a more systematic exploration of neural network classifier architectures and hyper-parameters, there is potential for even further improvements. Finally, we demonstrate that the time-frequency domain data representation allows us to efficiently trade-off a large reduction of model size and computational complexity for a very minor reduction in classification accuracy. This shows the potential for our classification approach to run on resource-constrained embedded and IoT devices.
Few-shot image classification is challenging due to the lack of ample samples in each class. Such a challenge becomes even tougher when the number of classes is very large, i.e., the large-class few-shot scenario. In this novel scenario, existing approaches do not perform well because they ignore confusable classes, namely similar classes that are difficult to distinguish from each other. These classes carry more information. In this paper, we propose a biased learning paradigm called Confusable Learning, which focuses more on confusable classes. Our method can be applied to mainstream meta-learning algorithms. Specifically, our method maintains a dynamically updating confusion matrix, which analyzes confusable classes in the dataset. Such a confusion matrix helps meta learners to emphasize on confusable classes. Comprehensive experiments on Omniglot, Fungi, and ImageNet demonstrate the efficacy of our method over state-of-the-art baselines.
Quantum Computing and especially Quantum Machine Learning, in a short period of time, has gained a lot of interest through research groups around the world. This can be seen in the increasing number of proposed models for pattern classification applying quantum principles to a certain degree. Despise the increasing volume of models, there is a void in testing these models on real datasets and not only on synthetic ones. The objective of this work is to classify patterns with binary attributes using a quantum classifier. Specially, we show results of a complete quantum classifier applied to image datasets. The experiments show favorable output while dealing with balanced classification problems as well as with imbalanced classes where the minority class is the most relevant. This is promising in medical areas, where usually the important class is also the minority class.
Financial institutions mostly deal with people. Therefore, characterizing different kinds of human behavior can greatly help institutions for improving their relation with customers and with regulatory offices. In many of such interactions, humans have some internal goals, and execute some actions within the financial system that lead them to achieve their goals. In this paper, we tackle these tasks as a behavior-traces classification task. An observer agent tries to learn characterizing other agents by observing their behavior when taking actions in a given environment. The other agents can be of several types and the goal of the observer is to identify the type of the other agent given a trace of observations. We present CABBOT, a learning technique that allows the agent to perform on-line classification of the type of planning agent whose behavior is observing. In this work, the observer agent has partial and noisy observability of the environment (state and actions of the other agents). In order to evaluate the performance of the learning technique, we have generated a domain-independent goal-based simulator of agents. We present experiments in several (both financial and non-financial) domains with promising results.
Patients with respiratory conditions typically exhibit adventitious respiratory sounds, such as wheezes. Wheeze events have variable duration. In this work we studied the influence of event duration on wheeze classification, namely how the creation of the non-wheeze class affected the classifiers' performance. First, we evaluated several classifiers on an open access respiratory sound database, with the best one reaching sensitivity and specificity values of 98% and 95%, respectively. Then, by changing one parameter in the design of the non-wheeze class, i.e., event duration, the best classifier only reached sensitivity and specificity values of 55% and 76%, respectively. These results demonstrate the importance of experimental design on the assessment of wheeze classification algorithms' performance.
Recently, deep neural networks have expanded the state-of-art in various scientific fields and provided solutions to long standing problems across multiple application domains. Nevertheless, they also suffer from weaknesses since their optimal performance depends on massive amounts of training data and the tuning of an extended number of parameters. As a countermeasure, some deep-forest methods have been recently proposed, as efficient and low-scale solutions. Despite that, these approaches simply employ label classification probabilities as induced features and primarily focus on traditional classification and regression tasks, leaving multi-output prediction under-explored. Moreover, recent work has demonstrated that tree-embeddings are highly representative, especially in structured output prediction. In this direction, we propose a novel deep tree-ensemble (DTE) model, where every layer enriches the original feature set with a representation learning component based on tree-embeddings. In this paper, we specifically focus on two structured output prediction tasks, namely multi-label classification and multi-target regression. We conducted experiments using multiple benchmark datasets and the obtained results confirm that our method provides superior results to state-of-the-art methods in both tasks.
In supervised learning, we often face with ambiguous (A) samples that are difficult to label even by domain experts. In this paper, we consider a binary classification problem in the presence of such A samples. This problem is substantially different from semi-supervised learning since unlabeled samples are not necessarily difficult samples. Also, it is different from 3-class classification with the positive (P), negative (N), and A classes since we do not want to classify test samples into the A class. Our proposed method extends binary classification with reject option, which trains a classifier and a rejector simultaneously using P and N samples based on the 0-1-$c$ loss with rejection cost $c$. More specifically, we propose to train a classifier and a rejector under the 0-1-$c$-$d$ loss using P, N, and A samples, where $d$ is the misclassification penalty for ambiguous samples. In our practical implementation, we use a convex upper bound of the 0-1-$c$-$d$ loss for computational tractability. Numerical experiments demonstrate that our method can successfully utilize the additional information brought by such A training data.