This paper introduces an improved generative model for statistical parametric speech synthesis (SPSS) based on WaveNet under a multi-task learning framework. Different from the original WaveNet model, the proposed Multi-task WaveNet employs the frame-level acoustic feature prediction as the secondary task and the external fundamental frequency prediction model for the original WaveNet can be removed. Therefore the improved WaveNet can generate high-quality speech waveforms only conditioned on linguistic features. Multi-task WaveNet can produce more natural and expressive speech by addressing the pitch prediction error accumulation issue and possesses more succinct inference procedures than the original WaveNet. Experimental results prove that the SPSS method proposed in this paper can achieve better performance than the state-of-the-art approach utilizing the original WaveNet in both objective and subjective preference tests.
This paper considers recommendation algorithm ensembles in a user-sensitive manner. Recently researchers have proposed various effective recommendation algorithms, which utilized different aspects of the data and different techniques. However, the "user skewed prediction" problem may exist for almost all recommendation algorithms -- algorithms with best average predictive accuracy may cover up that the algorithms may perform poorly for some part of users, which will lead to biased services in real scenarios. In this paper, we propose a user-sensitive ensemble method named "UREC" to address this issue. We first cluster users based on the recommendation predictions, then we use multi-task learning to learn the user-sensitive ensemble function for the users. In addition, to alleviate the negative effects of new user problem to clustering users, we propose an approximate approach based on a spectral relaxation. Experiments on real-world datasets demonstrate the superiority of our methods.
Performing more tasks in parallel is a typical feature of complex brains. These are characterized by the coexistence of excitatory and inhibitory synapses, whose percentage in mammals is measured to have a typical value of 20-30\%. Here we investigate parallel learning of more Boolean rules in neuronal networks. We find that multi-task learning results from the alternation of learning and forgetting of the individual rules. Interestingly, a fraction of 30\% inhibitory synapses optimizes the overall performance, carving a complex backbone supporting information transmission with a minimal shortest path length. We show that 30\% inhibitory synapses is the percentage maximizing the learning performance since it guarantees, at the same time, the network excitability necessary to express the response and the variability required to confine the employment of resources.
General image super-resolution techniques have difficulties in recovering detailed face structures when applying to low resolution face images. Recent deep learning based methods tailored for face images have achieved improved performance by jointly trained with additional task such as face parsing and landmark prediction. However, multi-task learning requires extra manually labeled data. Besides, most of the existing works can only generate relatively low resolution face images (e.g., $128\times128$), and their applications are therefore limited. In this paper, we introduce a novel SPatial Attention Residual Network (SPARNet) built on our newly proposed Face Attention Units (FAUs) for face super-resolution. Specifically, we introduce a spatial attention mechanism to the vanilla residual blocks. This enables the convolutional layers to adaptively bootstrap features related to the key face structures and pay less attention to those less feature-rich regions. This makes the training more effective and efficient as the key face structures only account for a very small portion of the face image. Visualization of the attention maps shows that our spatial attention network can capture the key face structures well even for very low resolution faces (e.g., $16\times16$). Quantitative comparisons on various kinds of metrics (including PSNR, SSIM, identity similarity, and landmark detection) demonstrate the superiority of our method over current state-of-the-arts. We further extend SPARNet with multi-scale discriminators, named as SPARNetHD, to produce high resolution results (i.e., $512\times512$). We show that SPARNetHD trained with synthetic data cannot only produce high quality and high resolution outputs for synthetically degraded face images, but also show good generalization ability to real world low quality face images.
Accuracy of many visiolinguistic tasks has benefited significantly from the application of vision-and-language(V&L) BERT. However, its application for the task of vision-and-language navigation (VLN) remains limited. One reason for this is the difficulty adapting the BERT architecture to the partially observable Markov decision process present in VLN, requiring history-dependent attention and decision making. In this paper we propose a recurrent BERT model that is time-aware for use in VLN. Specifically, we equip the BERT model with a recurrent function that maintains cross-modal state information for the agent. Through extensive experiments on R2R and REVERIE we demonstrate that our model can replace more complex encoder-decoder models to achieve state-of-the-art results. Moreover, our approach can be generalised to other transformer-based architectures, supports pre-training, and is capable of solving navigation and referring expression tasks simultaneously.
The data-driven nature of deep learning (DL) models for semantic segmentation requires a large number of pixel-level annotations. However, large-scale and fully labeled medical datasets are often unavailable for practical tasks. Recently, partially supervised methods have been proposed to utilize images with incomplete labels in the medical domain. To bridge the methodological gaps in partially supervised learning (PSL) under data scarcity, we propose Vicinal Labels Under Uncertainty (VLUU), a simple yet efficient framework utilizing the human structure similarity for partially supervised medical image segmentation. Motivated by multi-task learning and vicinal risk minimization, VLUU transforms the partially supervised problem into a fully supervised problem by generating vicinal labels. We systematically evaluate VLUU under the challenges of small-scale data, dataset shift, and class imbalance on two commonly used segmentation datasets for the tasks of chest organ segmentation and optic disc-and-cup segmentation. The experimental results show that VLUU can consistently outperform previous partially supervised models in these settings. Our research suggests a new research direction in label-efficient deep learning with partial supervision.
A flexible discriminative image denoiser is introduced in which multi-task learning methods are applied to a densoising FCN based on U-Net. The activations of the U-Net model are modified by affine transforms that are a learned function of conditioning inputs. The learning procedure for multiple noise types and levels involves applying a distribution of noise parameters during training to the conditioning inputs, with the same noise parameters applied to a noise generating layer at the input (similar to the approach taken in a denoising autoencoder). It is shown that this flexible denoising model achieves state of the art performance on images corrupted with Gaussian and Poisson noise. It has also been shown that this conditional training method can generalise a fixed noise level U-Net denoiser to a variety of noise levels.
In multi-task learning (MTL), we improve the performance of key machine learning algorithms by training various tasks jointly. When the number of tasks is large, modeling task structure can further refine the task relationship model. For example, often tasks can be grouped based on metadata, or via simple preprocessing steps like K-means. In this paper, we present our group structured latent-space multi-task learning model, which encourages group structured tasks defined by prior information. We use an alternating minimization method to learn the model parameters. Experiments are conducted on both synthetic and real-world datasets, showing competitive performance over single-task learning (where each group is trained separately) and other MTL baselines.
End-to-end automatic speech recognition (ASR) systems are increasingly popular due to their relative architectural simplicity and competitive performance. However, even though the average accuracy of these systems may be high, the performance on rare content words often lags behind hybrid ASR systems. To address this problem, second-pass rescoring is often applied leveraging upon language modeling. In this paper, we propose a second-pass system with multi-task learning, utilizing semantic targets (such as intent and slot prediction) to improve speech recognition performance. We show that our rescoring model trained with these additional tasks outperforms the baseline rescoring model, trained with only the language modeling task, by 1.4% on a general test and by 2.6% on a rare word test set in terms of word-error-rate relative (WERR). Our best ASR system with multi-task LM shows 4.6% WERR deduction compared with RNN Transducer only ASR baseline for rare words recognition.
Human Settlement Extent (HSE) and Local Climate Zone (LCZ) maps are both essential sources, e.g., for sustainable urban development and Urban Heat Island (UHI) studies. Remote sensing (RS)- and deep learning (DL)-based classification approaches play a significant role by providing the potential for global mapping. However, most of the efforts only focus on one of the two schemes, usually on a specific scale. This leads to unnecessary redundancies, since the learned features could be leveraged for both of these related tasks. In this letter, the concept of multi-task learning (MTL) is introduced to HSE regression and LCZ classification for the first time. We propose a MTL framework and develop an end-to-end Convolutional Neural Network (CNN), which consists of a backbone network for shared feature learning, attention modules for task-specific feature learning, and a weighting strategy for balancing the two tasks. We additionally propose to exploit HSE predictions as a prior for LCZ classification to enhance the accuracy. The MTL approach was extensively tested with Sentinel-2 data of 13 cities across the world. The results demonstrate that the framework is able to provide a competitive solution for both tasks.
The quality grading of mangoes is a crucial task for mango growers as it vastly affects their profit. However, until today, this process still relies on laborious efforts of humans, who are prone to fatigue and errors. To remedy this, the paper approaches the grading task with various convolutional neural networks (CNN), a tried-and-tested deep learning technology in computer vision. The models involved include Mask R-CNN (for background removal), the numerous past winners of the ImageNet challenge, namely AlexNet, VGGs, and ResNets; and, a family of self-defined convolutional autoencoder-classifiers (ConvAE-Clfs) inspired by the claimed benefit of multi-task learning in classification tasks. Transfer learning is also adopted in this work via utilizing the ImageNet pretrained weights. Besides elaborating on the preprocessing techniques, training details, and the resulting performance, we go one step further to provide explainable insights into the model's working with the help of saliency maps and principal component analysis (PCA). These insights provide a succinct, meaningful glimpse into the intricate deep learning black box, fostering trust, and can also be presented to humans in real-world use cases for reviewing the grading results.
Deep reinforcement learning (DRL) is a very active research area. However, several technical and scientific issues require to be addressed, amongst which we can mention data inefficiency, exploration-exploitation trade-off, and multi-task learning. Therefore, distributed modifications of DRL were introduced; agents that could be run on many machines simultaneously. In this article, we provide a survey of the role of the distributed approaches in DRL. We overview the state of the field, by studying the key research works that have a significant impact on how we can use distributed methods in DRL. We choose to overview these papers, from the perspective of distributed learning, and not the aspect of innovations in reinforcement learning algorithms. Also, we evaluate these methods on different tasks and compare their performance with each other and with single actor and learner agents.
We are interested in learning models of non-stationary environments, which can be framed as a multi-task learning problem. Model-free reinforcement learning algorithms can achieve good asymptotic performance in multi-task learning at a cost of extensive sampling, due to their approach, which requires learning from scratch. While model-based approaches are among the most data efficient learning algorithms, they still struggle with complex tasks and model uncertainties. Meta-reinforcement learning addresses the efficiency and generalization challenges on multi task learning by quickly leveraging the meta-prior policy for a new task. In this paper, we propose a meta-reinforcement learning approach to learn the dynamic model of a non-stationary environment to be used for meta-policy optimization later. Due to the sample efficiency of model-based learning methods, we are able to simultaneously train both the meta-model of the non-stationary environment and the meta-policy until dynamic model convergence. Then, the meta-learned dynamic model of the environment will generate simulated data for meta-policy optimization. Our experiment demonstrates that our proposed method can meta-learn the policy in a non-stationary environment with the data efficiency of model-based learning approaches while achieving the high asymptotic performance of model-free meta-reinforcement learning.
Given an image, a back-ground knowledge, and a set of questions about an object, human learners answer the questions very consistently regardless of question forms and semantic tasks. The current advancement in neural-network based Visual Question Answering (VQA), despite their impressive performance, cannot ensure such consistency due to identically distribution (i.i.d.) assumption. We propose a new model-agnostic logic constraint to tackle this issue by formulating a logically consistent loss in the multi-task learning framework as well as a data organisation called family-batch and hybrid-batch. To demonstrate usefulness of this proposal, we train and evaluate MAC-net based VQA machines with and without the proposed logically consistent loss and the proposed data organization. The experiments confirm that the proposed loss formulae and introduction of hybrid-batch leads to more consistency as well as better performance. Though the proposed approach is tested with MAC-net, it can be utilised in any other QA methods whenever the logical consistency between answers exist.
A major challenge in scene graph classification is that the appearance of objects and relations can be significantly different from one image to another. Previous works have addressed this by relational reasoning over all objects in an image or incorporating prior knowledge into classification. Unlike previous works, we do not consider separate models for perception and prior knowledge. Instead, we take a multi-task learning approach, where we implement the classification as an attention layer. This allows for the prior knowledge to emerge and propagate within the perception model. By enforcing the model also to represent the prior, we achieve a strong inductive bias. We show that our model can accurately generate commonsense knowledge and that the iterative injection of this knowledge to scene representations leads to significantly higher classification performance. Additionally, our model can be fine-tuned on external knowledge given as triples. When combined with self-supervised learning and with 1% of annotated images only, this gives more than 3% improvement in object classification, 26% in scene graph classification, and 36% in predicate prediction accuracy.
Recently, fully-convolutional one-stage networks have shown superior performance comparing to two-stage frameworks for instance segmentation as typically they can generate higher-quality mask predictions with less computation. In addition, their simple design opens up new opportunities for joint multi-task learning. In this paper, we demonstrate that adding a single classification layer for semantic segmentation, fully-convolutional instance segmentation networks can achieve state-of-the-art panoptic segmentation quality. This is made possible by our novel dynamic rank-1 convolution (DR1Conv), a novel dynamic module that can efficiently merge high-level context information with low-level detailed features which is beneficial for both semantic and instance segmentation. Importantly, the proposed new method, termed DR1Mask, can perform panoptic segmentation by adding a single layer. To our knowledge, DR1Mask is the first panoptic segmentation framework that exploits a shared feature map for both instance and semantic segmentation by considering both efficacy and efficiency. Not only our framework is much more efficient -- twice as fast as previous best two-branch approaches, but also the unified framework opens up opportunities for using the same context module to improve the performance for both tasks. As a byproduct, when performing instance segmentation alone, DR1Mask is 10% faster and 1 point in mAP more accurate than previous state-of-the-art instance segmentation network BlendMask. Code is available at: https://git.io/AdelaiDet
Dialog state tracking (DST) suffers from severe data sparsity. While many natural language processing (NLP) tasks benefit from transfer learning and multi-task learning, in dialog these methods are limited by the amount of available data and by the specificity of dialog applications. In this work, we successfully utilize non-dialog data from unrelated NLP tasks to train dialog state trackers. This opens the door to the abundance of unrelated NLP corpora to mitigate the data sparsity issue inherent to DST.
Most Siamese network-based trackers perform the tracking process without model update, and cannot learn targetspecific variation adaptively. Moreover, Siamese-based trackers infer the new state of tracked objects by generating axis-aligned bounding boxes, which contain extra background noise, and are unable to accurately estimate the rotation and scale transformation of moving objects, thus potentially reducing tracking performance. In this paper, we propose a novel Rotation-Scale Invariant Network (RSINet) to address the above problem. Our RSINet tracker consists of a target-distractor discrimination branch and a rotation-scale estimation branch, the rotation and scale knowledge can be explicitly learned by a multi-task learning method in an end-to-end manner. In addtion, the tracking model is adaptively optimized and updated under spatio-temporal energy control, which ensures model stability and reliability, as well as high tracking efficiency. Comprehensive experiments on OTB-100, VOT2018, and LaSOT benchmarks demonstrate that our proposed RSINet tracker yields new state-of-the-art performance compared with recent trackers, while running at real-time speed about 45 FPS.
DEtection TRansformer (DETR) for object detection reaches competitive performance compared with Faster R-CNN via a transformer encoder-decoder architecture. However, trained with scratch transformers, DETR needs large-scale training data and an extreme long training schedule even on COCO dataset. Inspired by the great success of pre-training transformers in natural language processing, we propose a novel pretext task named random query patch detection in Unsupervised Pre-training DETR (UP-DETR). Specifically, we randomly crop patches from the given image and then feed them as queries to the decoder. The model is pre-trained to detect these query patches from the input image. During the pre-training, we address two critical issues: multi-task learning and multi-query localization. (1) To trade off classification and localization preferences in the pretext task, we find that freezing the CNN backbone is the prerequisite for the success of pre-training transformers. (2) To perform multi-query localization, we develop UP-DETR with multi-query patch detection with attention mask. Besides, UP-DETR also provides a unified perspective for fine-tuning object detection and one-shot detection tasks. In our experiments, UP-DETR significantly boosts the performance of DETR with faster convergence and higher average precision on object detection, one-shot detection and panoptic segmentation. Code and pre-training models: https://github.com/dddzg/up-detr.
We present a novel deep learning framework that uses dynamic functional connectivity to simultaneously localize the language and motor areas of the eloquent cortex in brain tumor patients. Our method leverages convolutional layers to extract graph-based features from the dynamic connectivity matrices and a long-short term memory (LSTM) attention network to weight the relevant time points during classification. The final stage of our model employs multi-task learning to identify different eloquent subsystems. Our unique training strategy finds a shared representation between the cognitive networks of interest, which enables us to handle missing patient data. We evaluate our method on resting-state fMRI data from 56 brain tumor patients while using task fMRI activations as surrogate ground-truth labels for training and testing. Our model achieves higher localization accuracies than conventional deep learning approaches and can identify bilateral language areas even when trained on left-hemisphere lateralized cases. Hence, our method may ultimately be useful for preoperative mapping in tumor patients.