Ultrasound is a non-invasive imaging modality that can be conveniently used to classify suspicious breast nodules and potentially detect the onset of breast cancer. Recently, Convolutional Neural Networks (CNN) techniques have shown promising results in classifying ultrasound images of the breast into benign or malignant. However, CNN inference acts as a black-box model, and as such, its decision-making is not interpretable. Therefore, increasing effort has been dedicated to explaining this process, most notably through GRAD-CAM and other techniques that provide visual explanations into inner workings of CNNs. In addition to interpretation, these methods provide clinically important information, such as identifying the location for biopsy or treatment. In this work, we analyze how adversarial assaults that are practically undetectable may be devised to alter these importance maps dramatically. Furthermore, we will show that this change in the importance maps can come with or without altering the classification result, rendering them even harder to detect. As such, care must be taken when using these importance maps to shed light on the inner workings of deep learning. Finally, we utilize Multi-Task Learning (MTL) and propose a new network based on ResNet-50 to improve the classification accuracies. Our sensitivity and specificity is comparable to the state of the art results.
We introduce a novel deep learning based group activity recognition approach called the Pose Only Group Activity Recognition System (POGARS), designed to use only tracked poses of people to predict the performed group activity. In contrast to existing approaches for group activity recognition, POGARS uses 1D CNNs to learn spatiotemporal dynamics of individuals involved in a group activity and forgo learning features from pixel data. The proposed model uses a spatial and temporal attention mechanism to infer person-wise importance and multi-task learning for simultaneously performing group and individual action classification. Experimental results confirm that POGARS achieves highly competitive results compared to state-of-the-art methods on a widely used public volleyball dataset despite only using tracked pose as input. Further our experiments show by using pose only as input, POGARS has better generalization capabilities compared to methods that use RGB as input.
Superimposing visible watermarks on images provides a powerful weapon to cope with the copyright issue. Watermark removal techniques, which can strengthen the robustness of visible watermarks in an adversarial way, have attracted increasing research interest. Modern watermark removal methods perform watermark localization and background restoration simultaneously, which could be viewed as a multi-task learning problem. However, existing approaches suffer from incomplete detected watermark and degraded texture quality of restored background. Therefore, we design a two-stage multi-task network to address the above issues. The coarse stage consists of a watermark branch and a background branch, in which the watermark branch self-calibrates the roughly estimated mask and passes the calibrated mask to background branch to reconstruct the watermarked area. In the refinement stage, we integrate multi-level features to improve the texture quality of watermarked area. Extensive experiments on two datasets demonstrate the effectiveness of our proposed method.
Widespread and rapid dissemination of false news has made fact-checking an indispensable requirement. Given its time-consuming and labor-intensive nature, the task calls for an automated support to meet the demand. In this paper, we propose to leverage commonsense knowledge for the tasks of false news classification and check-worthy claim detection. Arguing that commonsense knowledge is a factor in human believability, we fine-tune the BERT language model with a commonsense question answering task and the aforementioned tasks in a multi-task learning environment. For predicting fine-grained false news types, we compare the proposed fine-tuned model's performance with the false news classification models on a public dataset as well as a newly collected dataset. We compare the model's performance with the single-task BERT model and a state-of-the-art check-worthy claim detection tool to evaluate the check-worthy claim detection. Our experimental analysis demonstrates that commonsense knowledge can improve performance in both tasks.
Numerous neural retrieval models have been proposed in recent years. These models learn to compute a ranking score between the given query and document. The majority of existing models are trained in pairwise fashion using human-judged labels directly without further calibration. The traditional pairwise schemes can be time-consuming and require pre-defined positive-negative document pairs for training, potentially leading to learning bias due to document distribution mismatch between training and test conditions. Some popular existing listwise schemes rely on the strong pre-defined probabilistic assumptions and stark difference between relevant and non-relevant documents for the given query, which may limit the model potential due to the low-quality or ambiguous relevance labels. To address these concerns, we turn to a physics-inspired ranking balance scheme and propose PoolRank, a pooling-based listwise learning framework. The proposed scheme has four major advantages: (1) PoolRank extracts training information from the best candidates at the local level based on model performance and relative ranking among abundant document candidates. (2) By combining four pooling-based loss components in a multi-task learning fashion, PoolRank calibrates the ranking balance for the partially relevant and the highly non-relevant documents automatically without costly human inspection. (3) PoolRank can be easily generalized to any neural retrieval model without requiring additional learnable parameters or model structure modifications. (4) Compared to pairwise learning and existing listwise learning schemes, PoolRank yields better ranking performance for all studied retrieval models while retaining efficient convergence rates.
Grasping in cluttered scenes has always been a great challenge for robots, due to the requirement of the ability to well understand the scene and object information. Previous works usually assume that the geometry information of the objects is available, or utilize a step-wise, multi-stage strategy to predict the feasible 6-DoF grasp poses. In this work, we propose to formalize the 6-DoF grasp pose estimation as a simultaneous multi-task learning problem. In a unified framework, we jointly predict the feasible 6-DoF grasp poses, instance semantic segmentation, and collision information. The whole framework is jointly optimized and end-to-end differentiable. Our model is evaluated on large-scale benchmarks as well as the real robot system. On the public dataset, our method outperforms prior state-of-the-art methods by a large margin (+4.08 AP). We also demonstrate the implementation of our model on a real robotic platform and show that the robot can accurately grasp target objects in cluttered scenarios with a high success rate. Project link: https://openbyterobotics.github.io/sscl
Given multiple non-convex objective functions and objective-specific weights, Chebyshev scalarization (CS) is a well-known approach to obtain an Exact Pareto Optimal (EPO), i.e., a solution on the Pareto front (PF) that intersects the ray defined by the inverse of the weights. First-order optimizers that use the CS formulation to find EPO solutions encounter practical problems of oscillations and stagnation that affect convergence. Moreover, when initialized with a PO solution, they do not guarantee a controlled trajectory that lies completely on the PF. These shortcomings lead to modeling limitations and computational inefficiency in multi-task learning (MTL) and multi-criteria decision-making (MCDM) methods that utilize CS for their underlying non-convex multi-objective optimization (MOO). To address these shortcomings, we design a new MOO method, EPO Search. We prove that EPO Search converges to an EPO solution and empirically illustrate its computational efficiency and robustness to initialization. When initialized on the PF, EPO Search can trace the PF and converge to the required EPO solution at a linear rate of convergence. Using EPO Search we develop new algorithms: PESA-EPO for approximating the PF in a posteriori MCDM, and GP-EPO for preference elicitation in interactive MCDM; experiments on benchmark datasets confirm their advantages over competing alternatives. EPO Search scales linearly with the number of decision variables which enables its use for training deep networks. Empirical results on real data from personalized medicine, e-commerce and hydrometeorology demonstrate the efficacy of EPO Search for deep MTL.
Masked language models (MLMs) are pre-trained with a denoising objective that is in a mismatch with the objective of downstream fine-tuning. We propose pragmatic masking and surrogate fine-tuning as two complementing strategies that exploit social cues to drive pre-trained representations toward a broad set of concepts useful for a wide class of social meaning tasks. We test our models on $15$ different Twitter datasets for social meaning detection. Our methods achieve $2.34\%$ $F_1$ over a competitive baseline, while outperforming domain-specific language models pre-trained on large datasets. Our methods also excel in few-shot learning: with only $5\%$ of training data (severely few-shot), our methods enable an impressive $68.54\%$ average $F_1$. The methods are also language agnostic, as we show in a zero-shot setting involving six datasets from three different languages.
Photo-realistic video portrait reenactment benefits virtual production and numerous VR/AR experiences. The task remains challenging as the reenacted expression should match the source while the lighting should be adjustable to new environments. We present a neural relighting and expression transfer technique to transfer the facial expressions from a source performer to a portrait video of a target performer while enabling dynamic relighting. Our approach employs 4D reflectance field learning, model-based facial performance capture and target-aware neural rendering. Specifically, given a short sequence of the target performer's OLAT, we apply a rendering-to-video translation network to first synthesize the OLAT result of new sequences with unseen expressions. We then design a semantic-aware facial normalization scheme along with a multi-frame multi-task learning strategy to encode the content, segmentation, and motion flows for reliably inferring the reflectance field. This allows us to simultaneously control facial expression and apply virtual relighting. Extensive experiments demonstrate that our technique can robustly handle challenging expressions and lighting environments and produce results at a cinematographic quality.
Understanding the emotional impact of movies has become important for affective movie analysis, ranking, and indexing. Methods for recognizing evoked emotions are usually trained on human annotated data. Concretely, viewers watch video clips and have to manually annotate the emotions they experienced while watching the videos. Then, the common practice is to aggregate the different annotations, by computing average scores or majority voting, and train and test models on these aggregated annotations. With this procedure a single aggregated evoked emotion annotation is obtained per each video. However, emotions experienced while watching a video are subjective: different individuals might experience different emotions. In this paper, we model the emotions evoked by videos in a different manner: instead of modeling the aggregated value we jointly model the emotions experienced by each viewer and the aggregated value using a multi-task learning approach. Concretely, we propose two deep learning architectures: a Single-Task (ST) architecture and a Multi-Task (MT) architecture. Our results show that the MT approach can more accurately model each viewer and the aggregated annotation when compared to methods that are directly trained on the aggregated annotations. Furthermore, our approach outperforms the current state-of-the-art results on the COGNIMUSE benchmark.
Deep neural networks set the state-of-the-art across many tasks in computer vision, but their generalization ability to image distortions is surprisingly fragile. In contrast, the mammalian visual system is robust to a wide range of perturbations. Recent work suggests that this generalization ability can be explained by useful inductive biases encoded in the representations of visual stimuli throughout the visual cortex. Here, we successfully leveraged these inductive biases with a multi-task learning approach: we jointly trained a deep network to perform image classification and to predict neural activity in macaque primary visual cortex (V1). We measured the out-of-distribution generalization abilities of our network by testing its robustness to image distortions. We found that co-training on monkey V1 data leads to increased robustness despite the absence of those distortions during training. Additionally, we showed that our network's robustness is very close to that of an Oracle network where parts of the architecture are directly trained on noisy images. Our results also demonstrated that the network's representations become more brain-like as their robustness improves. Using a novel constrained reconstruction analysis, we investigated what makes our brain-regularized network more robust. We found that our co-trained network is more sensitive to content than noise when compared to a Baseline network that we trained for image classification alone. Using DeepGaze-predicted saliency maps for ImageNet images, we found that our monkey co-trained network tends to be more sensitive to salient regions in a scene, reminiscent of existing theories on the role of V1 in the detection of object borders and bottom-up saliency. Overall, our work expands the promising research avenue of transferring inductive biases from the brain, and provides a novel analysis of the effects of our transfer.
With the rise in use of social media to promote branded products, the demand for effective influencer marketing has increased. Brands are looking for improved ways to identify valuable influencers among a vast catalogue; this is even more challenging with "micro-influencers", which are more affordable than mainstream ones but difficult to discover. In this paper, we propose a novel multi-task learning framework to improve the state of the art in micro-influencer ranking based on multimedia content. Moreover, since the visual congruence between a brand and influencer has been shown to be good measure of compatibility, we provide an effective visual method for interpreting our models' decisions, which can also be used to inform brands' media strategies. We compare with the current state-of-the-art on a recently constructed public dataset and we show significant improvement both in terms of accuracy and model complexity. The techniques for ranking and interpretation presented in this work can be generalised to arbitrary multimedia ranking tasks that have datasets with a similar structure.
By leveraging large amounts of product data collected across hundreds of live e-commerce websites, we construct 1000 unique classification tasks that share similarly-structured input data, comprised of both text and images. These classification tasks focus on learning the product hierarchy of different e-commerce websites, causing many of them to be correlated. Adopting a multi-modal transformer model, we solve these tasks in unison using multi-task learning (MTL). Extensive experiments are presented over an initial 100-task dataset to reveal best practices for "large-scale MTL" (i.e., MTL with more than 100 tasks). From these experiments, a final, unified methodology is derived, which is composed of both best practices and new proposals such as DyPa, a simple heuristic for automatically allocating task-specific parameters to tasks that could benefit from extra capacity. Using our large-scale MTL methodology, we successfully train a single model across all 1000 tasks in our dataset while using minimal task specific parameters, thereby showing that it is possible to extend several orders of magnitude beyond current efforts in MTL.
We present ENHANCE, an open dataset with multiple annotations to complement the existing ISIC and PH2 skin lesion classification datasets. This dataset contains annotations of visual ABC (asymmetry, border, colour) features from non-expert annotation sources: undergraduate students, crowd workers from Amazon MTurk and classic image processing algorithms. In this paper we first analyse the correlations between the annotations and the diagnostic label of the lesion, as well as study the agreement between different annotation sources. Overall we find weak correlations of non-expert annotations with the diagnostic label, and low agreement between different annotation sources. We then study multi-task learning (MTL) with the annotations as additional labels, and show that non-expert annotations can improve (ensembles of) state-of-the-art convolutional neural networks via MTL. We hope that our dataset can be used in further research into multiple annotations and/or MTL. All data and models are available on Github: https://github.com/raumannsr/ENHANCE.
Depth map super-resolution is a task with high practical application requirements in the industry. Existing color-guided depth map super-resolution methods usually necessitate an extra branch to extract high-frequency detail information from RGB image to guide the low-resolution depth map reconstruction. However, because there are still some differences between the two modalities, direct information transmission in the feature dimension or edge map dimension cannot achieve satisfactory result, and may even trigger texture copying in areas where the structures of the RGB-D pair are inconsistent. Inspired by the multi-task learning, we propose a joint learning network of depth map super-resolution (DSR) and monocular depth estimation (MDE) without introducing additional supervision labels. For the interaction of two subnetworks, we adopt a differentiated guidance strategy and design two bridges correspondingly. One is the high-frequency attention bridge (HABdg) designed for the feature encoding process, which learns the high-frequency information of the MDE task to guide the DSR task. The other is the content guidance bridge (CGBdg) designed for the depth map reconstruction process, which provides the content guidance learned from DSR task for MDE task. The entire network architecture is highly portable and can provide a paradigm for associating the DSR and MDE tasks. Extensive experiments on benchmark datasets demonstrate that our method achieves competitive performance. Our code and models are available at https://rmcong.github.io/proj_BridgeNet.html.
Constructing an embedding space for musical instrument sounds that can meaningfully represent new and unseen instruments is important for downstream music generation tasks such as multi-instrument synthesis and timbre transfer. The framework of Automatic Speaker Verification (ASV) provides us with architectures and evaluation methodologies for verifying the identities of unseen speakers, and these can be repurposed for the task of learning and evaluating a musical instrument sound embedding space that can support unseen instruments. Borrowing from state-of-the-art ASV techniques, we construct a musical instrument recognition model that uses a SincNet front-end, a ResNet architecture, and an angular softmax objective function. Experiments on the NSynth and RWC datasets show our model's effectiveness in terms of equal error rate (EER) for unseen instruments, and ablation studies show the importance of data augmentation and the angular softmax objective. Experiments also show the benefit of using a CQT-based filterbank for initializing SincNet over a Mel filterbank initialization. Further complementary analysis of the learned embedding space is conducted with t-SNE visualizations and probing classification tasks, which show that including instrument family labels as a multi-task learning target can help to regularize the embedding space and incorporate useful structure, and that meaningful information such as playing style, which was not included during training, is contained in the embeddings of unseen instruments.
This paper presents a PINN training framework that employs (1) pre-training steps that accelerates and improve the robustness of the training of physics-informed neural network with auxiliary data stored in point clouds, (2) a net-to-net knowledge transfer algorithm that improves the weight initialization of the neural network and (3) a multi-objective optimization algorithm that may improve the performance of a physical-informed neural network with competing constraints. We consider the training and transfer and multi-task learning of physics-informed neural network (PINN) as multi-objective problems where the physics constraints such as the governing equation, boundary conditions, thermodynamic inequality, symmetry, and invariant properties, as well as point cloud used for pre-training can sometimes lead to conflicts and necessitating the seek of the Pareto optimal solution. In these situations, weighted norms commonly used to handle multiple constraints may lead to poor performance, while other multi-objective algorithms may scale poorly with increasing dimensionality. To overcome this technical barrier, we adopt the concept of vectorized objective function and modify a gradient descent approach to handle the issue of conflicting gradients. Numerical experiments are compared the benchmark boundary value problems solved via PINN. The performance of the proposed paradigm is compared against the classical equal-weighted norm approach. Our numerical experiments indicate that the brittleness and lack of robustness demonstrated in some PINN implementations can be overcome with the proposed strategy.
Automatically mining sentiment tendency contained in natural language is a fundamental research to some artificial intelligent applications, where solutions alternate with challenges. Transfer learning and multi-task learning techniques have been leveraged to mitigate the supervision sparsity and collaborate multiple heterogeneous domains correspondingly. Recent years, the sensitive nature of users' private data raises another challenge for sentiment classification, i.e., data privacy protection. In this paper, we resort to federated learning for multiple domain sentiment classification under the constraint that the corpora must be stored on decentralized devices. In view of the heterogeneous semantics across multiple parties and the peculiarities of word embedding, we pertinently provide corresponding solutions. First, we propose a Knowledge Transfer Enhanced Private-Shared (KTEPS) framework for better model aggregation and personalization in federated sentiment classification. Second, we propose KTEPS$^\star$ with the consideration of the rich semantic and huge embedding size properties of word vectors, utilizing Projection-based Dimension Reduction (PDR) methods for privacy protection and efficient transmission simultaneously. We propose two federated sentiment classification scenes based on public benchmarks, and verify the superiorities of our proposed methods with abundant experimental investigations.
Analyzing and understanding hand information from multimedia materials like images or videos is important for many real world applications and remains active in research community. There are various works focusing on recovering hand information from single image, however, they usually solve a single task, for example, hand mask segmentation, 2D/3D hand pose estimation, or hand mesh reconstruction and perform not well in challenging scenarios. To further improve the performance of these tasks, we propose a novel Hand Image Understanding (HIU) framework to extract comprehensive information of the hand object from a single RGB image, by jointly considering the relationships between these tasks. To achieve this goal, a cascaded multi-task learning (MTL) backbone is designed to estimate the 2D heat maps, to learn the segmentation mask, and to generate the intermediate 3D information encoding, followed by a coarse-to-fine learning paradigm and a self-supervised learning strategy. Qualitative experiments demonstrate that our approach is capable of recovering reasonable mesh representations even in challenging situations. Quantitatively, our method significantly outperforms the state-of-the-art approaches on various widely-used datasets, in terms of diverse evaluation metrics.
In this paper, we propose a new framework for fine-grained emotion prediction in the text through emotion definition modeling. Our approach involves a multi-task learning framework that models definitions of emotions as an auxiliary task while being trained on the primary task of emotion prediction. We model definitions using masked language modeling and class definition prediction tasks. Our models outperform existing state-of-the-art for fine-grained emotion dataset GoEmotions. We further show that this trained model can be used for transfer learning on other benchmark datasets in emotion prediction with varying emotion label sets, domains, and sizes. The proposed models outperform the baselines on transfer learning experiments demonstrating the generalization capability of the models.