Ensuring traffic safety and mitigating accidents in modern driving is of paramount importance, and computer vision technologies have the potential to significantly contribute to this goal. This paper presents a multi-modal Vision Transformer for Driver Distraction Detection (termed ViT-DD), which incorporates inductive information from training signals related to both distraction detection and driver emotion recognition. Additionally, a self-learning algorithm is developed, allowing for the seamless integration of driver data without emotion labels into the multi-task training process of ViT-DD. Experimental results reveal that the proposed ViT-DD surpasses existing state-of-the-art methods for driver distraction detection by 6.5% and 0.9% on the SFDDD and AUCDD datasets, respectively.
Recently, self-learning methods based on user satisfaction metrics and contextual bandits have shown promising results to enable consistent improvements in conversational AI systems. However, directly targeting such metrics by off-policy bandit learning objectives often increases the risk of making abrupt policy changes that break the current user experience. In this study, we introduce a scalable framework for supporting fine-grained exploration targets for individual domains via user-defined constraints. For example, we may want to ensure fewer policy deviations in business-critical domains such as shopping, while allocating more exploration budget to domains such as music. Furthermore, we present a novel meta-gradient learning approach that is scalable and practical to address this problem. The proposed method adjusts constraint violation penalty terms adaptively through a meta objective that encourages balanced constraint satisfaction across domains. We conduct extensive experiments using data from a real-world conversational AI on a set of realistic constraint benchmarks. Based on the experimental results, we demonstrate that the proposed approach is capable of achieving the best balance between the policy value and constraint satisfaction rate.
Unsupervised video object segmentation aims to segment a target object in the video without a ground truth mask in the initial frame. This challenging task requires extracting features for the most salient common objects within a video sequence. This difficulty can be solved by using motion information such as optical flow, but using only the information between adjacent frames results in poor connectivity between distant frames and poor performance. To solve this problem, we propose a novel prototype memory network architecture. The proposed model effectively extracts the RGB and motion information by extracting superpixel-based component prototypes from the input RGB images and optical flow maps. In addition, the model scores the usefulness of the component prototypes in each frame based on a self-learning algorithm and adaptively stores the most useful prototypes in memory and discards obsolete prototypes. We use the prototypes in the memory bank to predict the next query frames mask, which enhances the association between distant frames to help with accurate mask prediction. Our method is evaluated on three datasets, achieving state-of-the-art performance. We prove the effectiveness of the proposed model with various ablation studies.
Runtime enforcement refers to the theories, techniques, and tools for enforcing correct behavior with respect to a formal specification of systems at runtime. In this paper, we are interested in techniques for constructing runtime enforcers for the concrete application domain of enforcing safety in AI. We discuss how safety is traditionally handled in the field of AI and how more formal guarantees on the safety of a self-learning agent can be given by integrating a runtime enforcer. We survey a selection of work on such enforcers, where we distinguish between approaches for discrete and continuous action spaces. The purpose of this paper is to foster a better understanding of advantages and limitations of different enforcement techniques, focusing on the specific challenges that arise due to their application in AI. Finally, we present some open challenges and avenues for future work.
Smart energy networks provide for an effective means to accommodate high penetrations of variable renewable energy sources like solar and wind, which are key for deep decarbonisation of energy production. However, given the variability of the renewables as well as the energy demand, it is imperative to develop effective control and energy storage schemes to manage the variable energy generation and achieve desired system economics and environmental goals. In this paper, we introduce a hybrid energy storage system composed of battery and hydrogen energy storage to handle the uncertainties related to electricity prices, renewable energy production and consumption. We aim to improve renewable energy utilisation and minimise energy costs and carbon emissions while ensuring energy reliability and stability within the network. To achieve this, we propose a multi-agent deep deterministic policy gradient approach, which is a deep reinforcement learning-based control strategy to optimise the scheduling of the hybrid energy storage system and energy demand in real-time. The proposed approach is model-free and does not require explicit knowledge and rigorous mathematical models of the smart energy network environment. Simulation results based on real-world data show that: (i) integration and optimised operation of the hybrid energy storage system and energy demand reduces carbon emissions by 78.69%, improves cost savings by 23.5% and renewable energy utilisation by over 13.2% compared to other baseline models and (ii) the proposed algorithm outperforms the state-of-the-art self-learning algorithms like deep-Q network.
Collection of real world annotations for training semantic segmentation models is an expensive process. Unsupervised domain adaptation (UDA) tries to solve this problem by studying how more accessible data such as synthetic data can be used to train and adapt models to real world images without requiring their annotations. Recent UDA methods applies self-learning by training on pixel-wise classification loss using a student and teacher network. In this paper, we propose the addition of a consistency regularization term to semi-supervised UDA by modelling the inter-pixel relationship between elements in networks' output. We demonstrate the effectiveness of the proposed consistency regularization term by applying it to the state-of-the-art DAFormer framework and improving mIoU19 performance on the GTA5 to Cityscapes benchmark by 0.8 and mIou16 performance on the SYNTHIA to Cityscapes benchmark by 1.2.
Images acquired during underwater activities suffer from environmental properties of the water, such as turbidity and light attenuation. These phenomena cause color distortion, blurring, and contrast reduction. In addition, irregular ambient light distribution causes color channel unbalance and regions with high-intensity pixels. Recent works related to underwater image enhancement, and based on deep learning approaches, tackle the lack of paired datasets generating synthetic ground-truth. In this paper, we present a self-supervised learning methodology for underwater image enhancement based on deep learning that requires no paired datasets. The proposed method estimates the degradation present in underwater images. Besides, an autoencoder reconstructs this image, and its output image is degraded using the estimated degradation information. Therefore, the strategy replaces the output image with the degraded version in the loss function during the training phase. This procedure \textit{misleads} the neural network that learns to compensate the additional degradation. As a result, the reconstructed image is an enhanced version of the input image. Also, the algorithm presents an attention module to reduce high-intensity areas generated in enhanced images by color channel unbalances and outlier regions. Furthermore, the proposed methodology requires no ground-truth. Besides, only real underwater images were used to train the neural network, and the results indicate the effectiveness of the method in terms of color preservation, color cast reduction, and contrast improvement.
Exploiting unmanned aerial vehicles (UAVs) to execute tasks is gaining growing popularity recently. To solve the underlying task scheduling problem, the deep reinforcement learning (DRL) based methods demonstrate notable advantage over the conventional heuristics as they rely less on hand-engineered rules. However, their decision space will become prohibitively huge as the problem scales up, thus deteriorating the computation efficiency. To alleviate this issue, we propose a double-level deep reinforcement learning (DL-DRL) approach based on a divide and conquer framework (DCF), where we decompose the task scheduling of multi-UAV into task allocation and route planning. Particularly, we design an encoder-decoder structured policy network in our upper-level DRL model to allocate the tasks to different UAVs, and we exploit another attention based policy network in our lower-level DRL model to construct the route for each UAV, with the objective to maximize the number of executed tasks given the maximum flight distance of the UAV. To effectively train the two models, we design an interactive training strategy (ITS), which includes pre-training, intensive training and alternate training. Experimental results show that our DL-DRL performs favorably against the learning-based and conventional baselines including the OR-Tools, in terms of solution quality and computation efficiency. We also verify the generalization performance of our approach by applying it to larger sizes of up to 1000 tasks. Moreover, we also show via an ablation study that our ITS can help achieve a balance between the performance and training efficiency.
Cyber Threat Intelligence (CTI) is information describing threat vectors, vulnerabilities, and attacks and is often used as training data for AI-based cyber defense systems such as Cybersecurity Knowledge Graphs (CKG). There is a strong need to develop community-accessible datasets to train existing AI-based cybersecurity pipelines to efficiently and accurately extract meaningful insights from CTI. We have created an initial unstructured CTI corpus from a variety of open sources that we are using to train and test cybersecurity entity models using the spaCy framework and exploring self-learning methods to automatically recognize cybersecurity entities. We also describe methods to apply cybersecurity domain entity linking with existing world knowledge from Wikidata. Our future work will survey and test spaCy NLP tools and create methods for continuous integration of new information extracted from text.
Deep Learning-based Unsupervised Salient Object Detection (USOD) mainly relies on the noisy saliency pseudo labels that have been generated from traditional handcraft methods or pre-trained networks. To cope with the noisy labels problem, a class of methods focus on only easy samples with reliable labels but ignore valuable knowledge in hard samples. In this paper, we propose a novel USOD method to mine rich and accurate saliency knowledge from both easy and hard samples. First, we propose a Confidence-aware Saliency Distilling (CSD) strategy that scores samples conditioned on samples' confidences, which guides the model to distill saliency knowledge from easy samples to hard samples progressively. Second, we propose a Boundary-aware Texture Matching (BTM) strategy to refine the boundaries of noisy labels by matching the textures around the predicted boundary. Extensive experiments on RGB, RGB-D, RGB-T, and video SOD benchmarks prove that our method achieves state-of-the-art USOD performance.
Recent advances in big/foundation models reveal a promising path for deep learning, where the roadmap steadily moves from big data to big models to (the newly-introduced) big learning. Specifically, the big learning exhaustively exploits the information inherent in its large-scale complete/incomplete training data, by simultaneously modeling many/all joint/conditional/marginal data distributions across potentially diverse domains, with one universal foundation model. We reveal that big learning ($i$) underlies most existing foundation models, ($ii$) is equipped with extraordinary flexibilities for complete/incomplete training data and trustworthy data tasks, ($iii$) is capable of delivering all joint/conditional/marginal data capabilities with one universal model, and ($iv$) unifies conventional machine learning paradigms and enables their flexible cooperations, manifested as a universal learning paradigm. Diverse experiments are carried out to validate the effectiveness of the presented big learning.
Automatic log file analysis enables early detection of relevant incidents such as system failures. In particular, self-learning anomaly detection techniques capture patterns in log data and subsequently report unexpected log event occurrences to system operators without the need to provide or manually model anomalous scenarios in advance. Recently, an increasing number of approaches leveraging deep learning neural networks for this purpose have been presented. These approaches have demonstrated superior detection performance in comparison to conventional machine learning techniques and simultaneously resolve issues with unstable data formats. However, there exist many different architectures for deep learning and it is non-trivial to encode raw and unstructured log data to be analyzed by neural networks. We therefore carry out a systematic literature review that provides an overview of deployed models, data pre-processing mechanisms, anomaly detection techniques, and evaluations. The survey does not quantitatively compare existing approaches but instead aims to help readers understand relevant aspects of different model architectures and emphasizes open issues for future work.
Although the vision-and-language pretraining (VLP) equipped cross-modal image-text retrieval (ITR) has achieved remarkable progress in the past two years, it suffers from a major drawback: the ever-increasing size of VLP models restricts its deployment to real-world search scenarios (where the high latency is unacceptable). To alleviate this problem, we present a novel plug-in dynamic contrastive distillation (DCD) framework to compress the large VLP models for the ITR task. Technically, we face the following two challenges: 1) the typical uni-modal metric learning approach is difficult to directly apply to the cross-modal tasks, due to the limited GPU memory to optimize too many negative samples during handling cross-modal fusion features. 2) it is inefficient to static optimize the student network from different hard samples, which have different effects on distillation learning and student network optimization. We try to overcome these challenges from two points. First, to achieve multi-modal contrastive learning, and balance the training costs and effects, we propose to use a teacher network to estimate the difficult samples for students, making the students absorb the powerful knowledge from pre-trained teachers, and master the knowledge from hard samples. Second, to dynamic learn from hard sample pairs, we propose dynamic distillation to dynamically learn samples of different difficulties, from the perspective of better balancing the difficulty of knowledge and students' self-learning ability. We successfully apply our proposed DCD strategy to two state-of-the-art vision-language pretrained models, i.e. ViLT and METER. Extensive experiments on MS-COCO and Flickr30K benchmarks show the effectiveness and efficiency of our DCD framework. Encouragingly, we can speed up the inference at least 129$\times$ compared to the existing ITR models.
Recent studies have shown great promise in applying graph neural networks for multivariate time series forecasting, where the interactions of time series are described as a graph structure and the variables are represented as the graph nodes. Along this line, existing methods usually assume that the graph structure (or the adjacency matrix), which determines the aggregation manner of graph neural network, is fixed either by definition or self-learning. However, the interactions of variables can be dynamic and evolutionary in real-world scenarios. Furthermore, the interactions of time series are quite different if they are observed at different time scales. To equip the graph neural network with a flexible and practical graph structure, in this paper, we investigate how to model the evolutionary and multi-scale interactions of time series. In particular, we first provide a hierarchical graph structure cooperated with the dilated convolution to capture the scale-specific correlations among time series. Then, a series of adjacency matrices are constructed under a recurrent manner to represent the evolving correlations at each layer. Moreover, a unified neural network is provided to integrate the components above to get the final prediction. In this way, we can capture the pair-wise correlations and temporal dependency simultaneously. Finally, experiments on both single-step and multi-step forecasting tasks demonstrate the superiority of our method over the state-of-the-art approaches.
Transformers have been widely applied in text classification. Unfortunately, real-world data contain anomalies and noisy labels that cause challenges for state-of-art Transformers. This paper proposes Protoformer, a novel self-learning framework for Transformers that can leverage problematic samples for text classification. Protoformer features a selection mechanism for embedding samples that allows us to efficiently extract and utilize anomalies prototypes and difficult class prototypes. We demonstrated such capabilities on datasets with diverse textual structures (e.g., Twitter, IMDB, ArXiv). We also applied the framework to several models. The results indicate that Protoformer can improve current Transformers in various empirical settings.
This paper proposes a novel pixel-level distribution regularization scheme (DRSL) for self-supervised domain adaptation of semantic segmentation. In a typical setting, the classification loss forces the semantic segmentation model to greedily learn the representations that capture inter-class variations in order to determine the decision (class) boundary. Due to the domain shift, this decision boundary is unaligned in the target domain, resulting in noisy pseudo labels adversely affecting self-supervised domain adaptation. To overcome this limitation, along with capturing inter-class variation, we capture pixel-level intra-class variations through class-aware multi-modal distribution learning (MMDL). Thus, the information necessary for capturing the intra-class variations is explicitly disentangled from the information necessary for inter-class discrimination. Features captured thus are much more informative, resulting in pseudo-labels with low noise. This disentanglement allows us to perform separate alignments in discriminative space and multi-modal distribution space, using cross-entropy based self-learning for the former. For later, we propose a novel stochastic mode alignment method, by explicitly decreasing the distance between the target and source pixels that map to the same mode. The distance metric learning loss, computed over pseudo-labels and backpropagated from multi-modal modeling head, acts as the regularizer over the base network shared with the segmentation head. The results from comprehensive experiments on synthetic to real domain adaptation setups, i.e., GTA-V/SYNTHIA to Cityscapes, show that DRSL outperforms many existing approaches (a minimum margin of 2.3% and 2.5% in mIoU for SYNTHIA to Cityscapes).
Vision transformers have generated significant interest in the computer vision community because of their flexibility in exploiting contextual information, whether it is sharply confined local, or long range global. However, they are known to be data hungry. This has motivated the research in self-supervised transformer pretraining, which does not need to decode the semantic information conveyed by labels to link it to the image properties, but rather focuses directly on extracting a concise representation of the image data that reflects the notion of similarity, and is invariant to nuisance factors. The key vehicle for the self-learning process used by the majority of self-learning methods is the generation of multiple views of the training data and the creation of pretext tasks which use these views to define the notion of image similarity, and data integrity. However, this approach lacks the natural propensity to extract contextual information. We propose group masked model learning (GMML), a self-supervised learning (SSL) mechanism for pretraining vision transformers with the ability to extract the contextual information present in all the concepts in an image. GMML achieves this by manipulating randomly groups of connected tokens, ensuingly covering a meaningful part of a semantic concept, and then recovering the hidden semantic information from the visible part of the concept. GMML implicitly introduces a novel data augmentation process. Unlike most of the existing SSL approaches, GMML does not require momentum encoder, nor rely on careful implementation details such as large batches and gradient stopping, which are all artefacts of most of the current self-supervised learning techniques. The source code is publicly available for the community to train on bigger corpora: https://github.com/Sara-Ahmed/GMML.
Predicted inter-chain residue-residue contacts can be used to build the quaternary structure of protein complexes from scratch. However, only a small number of methods have been developed to reconstruct protein quaternary structures using predicted inter-chain contacts. Here, we present an agent-based self-learning method based on deep reinforcement learning (DRLComplex) to build protein complex structures using inter-chain contacts as distance constraints. We rigorously tested DRLComplex on two standard datasets of homodimeric and heterodimeric protein complexes (i.e., the CASP-CAPRI homodimer and Std_32 heterodimer datasets) using both true and predicted interchain contacts as inputs. Utilizing true contacts as input, DRLComplex achieved high average TM-scores of 0.9895 and 0.9881 and a low average interface RMSD (I_RMSD) of 0.2197 and 0.92 on the two datasets, respectively. When predicted contacts are used, the method achieves TM-scores of 0.73 and 0.76 for homodimers and heterodimers, respectively. Our experiments find that the accuracy of reconstructed quaternary structures depends on the accuracy of the contact predictions. Compared to other optimization methods for reconstructing quaternary structures from inter-chain contacts, DRLComplex performs similar to an advanced gradient descent method and better than a Markov Chain Monte Carlo simulation method and a simulated annealing-based method, validating the effectiveness of DRLComplex for quaternary reconstruction of protein complexes.
Traditional vision based Automated Optical Inspection (referred to as AOI in paper) systems present multiple challenges in factory settings including inability to scale across multiple product lines, requirement of vendor programming expertise, little tolerance to variations and lack of cloud connectivity for aggregated insights. The lack of flexibility in these systems presents a unique opportunity for a deep learning based AOI system specifically for factory automation. The proposed solution, VizInspect pro is a generic computer vision based AOI solution built on top of Leo - An edge AI platform. Innovative features that overcome challenges of traditional vision systems include deep learning based image analysis which combines the power of self-learning with high speed and accuracy, an intuitive user interface to configure inspection profiles in minutes without ML or vision expertise and the ability to solve complex inspection challenges while being tolerant to deviations and unpredictable defects. This solution has been validated by multiple external enterprise customers with confirmed value propositions. In this paper we show you how this solution and platform solved problems around model development, deployment, scaling multiple inferences and visualizations.
Self-learning paradigms in large-scale conversational AI agents tend to leverage user feedback in bridging between what they say and what they mean. However, such learning, particularly in Markov-based query rewriting systems have far from addressed the impact of these models on future training where successive feedback is inevitably contingent on the rewrite itself, especially in a continually updating environment. In this paper, we explore the consequences of this inherent lack of self-awareness towards impairing the model performance, ultimately resulting in both Type I and II errors over time. To that end, we propose augmenting the Markov Graph construction with a superposition-based adjacency matrix. Here, our method leverages an induced stochasticity to reactively learn a locally-adaptive decision boundary based on the performance of the individual rewrites in a bi-variate beta setting. We also surface a data augmentation strategy that leverages template-based generation in abridging complex conversation hierarchies of dialogs so as to simplify the learning process. All in all, we demonstrate that our self-aware model improves the overall PR-AUC by 27.45%, achieves a relative defect reduction of up to 31.22%, and is able to adapt quicker to changes in global preferences across a large number of customers.