In recent years, mining the knowledge from asynchronous sequences by Hawkes process is a subject worthy of continued attention, and Hawkes processes based on the neural network have gradually become the most hotly researched fields, especially based on the recurrence neural network (RNN). However, these models still contain some inherent shortcomings of RNN, such as vanishing and exploding gradient and long-term dependency problems. Meanwhile, Transformer based on self-attention has achieved great success in sequential modeling like text processing and speech recognition. Although the Transformer Hawkes process (THP) has gained huge performance improvement, THPs do not effectively utilize the temporal information in the asynchronous events, for these asynchronous sequences, the event occurrence instants are as important as the types of events, while conventional THPs simply convert temporal information into position encoding and add them as the input of transformer. With this in mind, we come up with a new kind of Transformer-based Hawkes process model, Temporal Attention Augmented Transformer Hawkes Process (TAA-THP), we modify the traditional dot-product attention structure, and introduce the temporal encoding into attention structure. We conduct numerous experiments on a wide range of synthetic and real-life datasets to validate the performance of our proposed TAA-THP model, significantly improvement compared with existing baseline models on the different measurements is achieved, including log-likelihood on the test dataset, and prediction accuracies of event types and occurrence times. In addition, through the ablation studies, we vividly demonstrate the merit of introducing additional temporal attention by comparing the performance of the model with and without temporal attention.
Audio-visual automatic speech recognition (AV-ASR) extends the speech recognition by introducing the video modality. In particular, the information contained in the motion of the speaker's mouth is used to augment the audio features. The video modality is traditionally processed with a 3D convolutional neural network (e.g. 3D version of VGG). Recently, image transformer networks arXiv:2010.11929 demonstrated the ability to extract rich visual features for the image classification task. In this work, we propose to replace the 3D convolution with a video transformer video feature extractor. We train our baselines and the proposed model on a large scale corpus of the YouTube videos. Then we evaluate the performance on a labeled subset of YouTube as well as on the public corpus LRS3-TED. Our best model video-only model achieves the performance of 34.9% WER on YTDEV18 and 19.3% on LRS3-TED which is a 10% and 9% relative improvements over the convolutional baseline. We achieve the state of the art performance of the audio-visual recognition on the LRS3-TED after fine-tuning our model (1.6% WER).
Video understanding requires reasoning at multiple spatiotemporal resolutions -- from short fine-grained motions to events taking place over longer durations. Although transformer architectures have recently advanced the state-of-the-art, they have not explicitly modelled different spatiotemporal resolutions. To this end, we present Multiview Transformers for Video Recognition (MTV). Our model consists of separate encoders to represent different views of the input video with lateral connections to fuse information across views. We present thorough ablation studies of our model and show that MTV consistently performs better than single-view counterparts in terms of accuracy and computational cost across a range of model sizes. Furthermore, we achieve state-of-the-art results on five standard datasets, and improve even further with large-scale pretraining. We will release code and pretrained checkpoints.
We propose an end-to-end image compression and analysis model with Transformers, targeting to the cloud-based image classification application. Instead of placing an existing Transformer-based image classification model directly after an image codec, we aim to redesign the Vision Transformer (ViT) model to perform image classification from the compressed features and facilitate image compression with the long-term information from the Transformer. Specifically, we first replace the patchify stem (i.e., image splitting and embedding) of the ViT model with a lightweight image encoder modelled by a convolutional neural network. The compressed features generated by the image encoder are injected convolutional inductive bias and are fed to the Transformer for image classification bypassing image reconstruction. Meanwhile, we propose a feature aggregation module to fuse the compressed features with the selected intermediate features of the Transformer, and feed the aggregated features to a deconvolutional neural network for image reconstruction. The aggregated features can obtain the long-term information from the self-attention mechanism of the Transformer and improve the compression performance. The rate-distortion-accuracy optimization problem is finally solved by a two-step training strategy. Experimental results demonstrate the effectiveness of the proposed model in both the image compression and the classification tasks.
Language is not only used to inform. We often seek to persuade by arguing in favor of a particular view. Persuasion raises a number of challenges for classical accounts of belief updating, as information cannot be taken at face value. How should listeners account for a speaker's "hidden agenda" when incorporating new information? Here, we extend recent probabilistic models of recursive social reasoning to allow for persuasive goals and show that our model provides a new pragmatic explanation for why weakly favorable arguments may backfire, a phenomenon known as the weak evidence effect. Critically, our model predicts a relationship between belief updating and speaker expectations: weak evidence should only backfire when speakers are expected to act under persuasive goals, implying the absence of stronger evidence. We introduce a simple experimental paradigm called the Stick Contest to measure the extent to which the weak evidence effect depends on speaker expectations, and show that a pragmatic listener model accounts for the empirical data better than alternative models. Our findings suggest potential avenues for rational models of social reasoning to further illuminate decision-making phenomena.
Several studies have shown the ability of natural gradient descent to minimize the objective function more efficiently than ordinary gradient descent based methods. However, the bottleneck of this approach for training deep neural networks lies in the prohibitive cost of solving a large dense linear system corresponding to the Fisher Information Matrix (FIM) at each iteration. This has motivated various approximations of either the exact FIM or the empirical one. The most sophisticated of these is KFAC, which involves a Kronecker-factored block diagonal approximation of the FIM. With only a slight additional cost, a few improvements of KFAC from the standpoint of accuracy are proposed. The common feature of the four novel methods is that they rely on a direct minimization problem, the solution of which can be computed via the Kronecker product singular value decomposition technique. Experimental results on the three standard deep auto-encoder benchmarks showed that they provide more accurate approximations to the FIM. Furthermore, they outperform KFAC and state-of-the-art first-order methods in terms of optimization speed.
In recent years, machine learning neural network has penetrated deeply into people's life. As the price of convenience, people's private information also has the risk of disclosure. The "right to be forgotten" was introduced in a timely manner, stipulating that individuals have the right to withdraw their consent from personal information processing activities based on their consent. To solve this problem, machine unlearning is proposed, which allows the model to erase all memory of private information. Previous studies, including retraining and incremental learning to update models, often take up extra storage space or are difficult to apply to neural networks. Our method only needs to make a small perturbation of the weight of the target model and make it iterate in the direction of the model trained with the remaining data subset until the contribution of the unlearning data to the model is completely eliminated. In this paper, experiments on five datasets prove the effectiveness of our method for machine unlearning, and our method is 15 times faster than retraining.
Despite the recent advances in multi-task learning of dense prediction problems, most methods rely on expensive labelled datasets. In this paper, we present a label efficient approach and look at jointly learning of multiple dense prediction tasks on partially annotated data, which we call multi-task partially-supervised learning. We propose a multi-task training procedure that successfully leverages task relations to supervise its multi-task learning when data is partially annotated. In particular, we learn to map each task pair to a joint pairwise task-space which enables sharing information between them in a computationally efficient way through another network conditioned on task pairs, and avoids learning trivial cross-task relations by retaining high-level information about the input image. We rigorously demonstrate that our proposed method effectively exploits the images with unlabelled tasks and outperforms existing semi-supervised learning approaches and related methods on three standard benchmarks.
In recent years, segmentation methods based on deep convolutional neural networks (CNNs) have made state-of-the-art achievements for many medical analysis tasks. However, most of these approaches improve performance by optimizing the structure or adding new functional modules of the U-Net, which ignoring the complementation and fusion of the coarse-grained and fine-grained semantic information. To solve the above problems, we propose a medical image segmentation framework called progressive learning network (PL-Net), which includes internal progressive learning (IPL) and external progressive learning (EPL). PL-Net has the following advantages: (1) IPL divides feature extraction into two "steps", which can mix different size receptive fields and capture semantic information from coarse to fine granularity without introducing additional parameters; (2) EPL divides the training process into two "stages" to optimize parameters, and realizes the fusion of coarse-grained information in the previous stage and fine-grained information in the latter stage. We evaluate our method in different medical image analysis tasks, and the results show that the segmentation performance of PL-Net is better than the state-of-the-art methods of U-Net and its variants.
Here, we present a new approach to generate smooth control sequences in Model Predictive Path Integral control (MPPI) tasks without any additional smoothing algorithms. Our method effectively alleviates the chattering in sampling, while the information theoretic derivation of MPPI remains the same. We demonstrated the proposed method in a challenging autonomous driving task with quantitative evaluation of different algorithms. A neural network vehicle model for estimating system dynamics under varying road friction conditions is also presented. Our video can be found at: \url{https://youtu.be/o3Nmi0UJFqg}.