Vision, Learning and Control University of Southampton Southampton, UK Abstract In this paper, we present a new approach to computing the generalisation performance assuming that the distribution of risks, ρ (r), for a learning scenario is known. This allows us to compute the expected error of a learning machine using empirical risk minimisation. We show that it is possible to obtain results for both classification and regression. We show a critical quantity in determining the generalisation performance is the power-law behaviour of ρ ( r) around its minimum value. We compute ρ ( r) for the case of all Boolean functions and for the perceptron. We start with a simplistic analysis but then do a more formal one later on. We show that the simplistic results are qualitatively correct and provide a good approximation to the actual results if we replace the true training set size with an approximate training set size. Keywords: Generalisation, Learning Theory 1. Introduction Traditional computational learning theory aims to eliminate all rules that do not correctly explain the data. A rule can be thought of as a fixed set of parameters of a learning machine; more formally, a hypothesis. This process relies on the idea that rules with poor generalisation performance (high risk) will, with high probability, make errors on a sufficiently large randomly chosen training data set (Vapnik and Chervonenkis, 1971; Valiant, 1984; Baum and Haussler, 1989; Blumer et al., 1989; Haussler, 1992; Vapnik, 1992). Suppose there exists a mechanism for selecting a rule from the subset of rules that have the lowest errors on the training set. Then, there is a very small probability that any of the selected rules has a high risk. However, this crucially depends on there being effectively a finite number of hypotheses, otherwise, there could still be a high-risk set of parameters which by chance did well on the particular training set. In the case where the learning machine has a continuous parameter space (so that the dimensionality of the space is uncountably infinite), we consider the effective size of the hypothesis space to be the Vapnik-Chervonenkis (VC) dimension. The VC dimension measures the number of possible ways in which the machine can give different outputs to a finite number of training examples (Vapnik and Chervonenkis, 1971). This effective size or capacity lies at the heart of conventional computational learning theory. By limiting the capacity we can obtain stronger bounds on the generalisation performance. In this paper, we challenge this traditional approach.

We present a simple self-training method that achieves 87.4% top-1 accuracy on ImageNet, which is 1.0% better than the state-of-the-art model that requires 3.5B weakly labeled Instagram images. On robustness test sets, it improves ImageNet-A top-1 accuracy from 16.6% to 74.2%, reduces ImageNet-C mean corruption error from 45.7 to 31.2, and reduces ImageNet-P mean flip rate from 27.8 to 16.1. To achieve this result, we first train an EfficientNet model on labeled ImageNet images and use it as a teacher to generate pseudo labels on 300M unlabeled images. We then train a larger EfficientNet as a student model on the combination of labeled and pseudo labeled images. We iterate this process by putting back the student as the teacher. During the generation of the pseudo labels, the teacher is not noised so that the pseudo labels are as good as possible. But during the learning of the student, we inject noise such as data augmentation, dropout, stochastic depth to the student so that the noised student is forced to learn harder from the pseudo labels.

Visual objects are composed of a recursive hierarchy of perceptual wholes and parts, whose properties, such as shape, reflectance, and color, constitute a hierarchy of intrinsic causal factors of object appearance. However, object appearance is the compositional consequence of both an object's intrinsic and extrinsic causal factors, where the extrinsic causal factors are related to illumination, and imaging conditions. Therefore, this paper proposes a unified tensor model of wholes and parts, and introduces a compositional hierarchical tensor factorization that disentangles the hierarchical causal structure of object image formation, and subsumes multilinear block tensor decomposition as a special case. The resulting object representation is an interpretable combinatorial choice of wholes' and parts' representations that renders object recognition robust to occlusion and reduces training data requirements. We demonstrate ourapproach in the context of face recognition by training on an extremely reduced dataset of synthetic images, and report encouragingface verification results on two datasets - the Freiburg dataset, andthe Labeled Face in the Wild (LFW) dataset consisting of real world images, thus, substantiating the suitability of our approach for data starved domains.

Time series modeling has attracted extensive research efforts; however, achieving both reliable efficiency and interpretability from a unified model still remains a challenging problem. Among the literature, shapelets offer interpretable and explanatory insights in the classification tasks, while most existing works ignore the differing representative power at different time slices, as well as (more importantly) the evolution pattern of shapelets. In this paper, we propose to extract time-aware shapelets by designing a two-level timing factor. Moreover, we define and construct the shapelet evolution graph, which captures how shapelets evolve over time and can be incorporated into the time series embeddings by graph embedding algorithms. To validate whether the representations obtained in this way can be applied effectively in various scenarios, we conduct experiments based on three public time series datasets, and two real-world datasets from different domains. Experimental results clearly show the improvements achieved by our approach compared with 17 state-of-the-art baselines.

Good data stewardship requires removal of data at the request of the data's owner. This raises the question if and how a trained machine-learning model, which implicitly stores information about its training data, should be affected by such a removal request. Is it possible to "remove" data from a machine-learning model? We study this problem by defining certified removal: a very strong theoretical guarantee that a model from which data is removed cannot be distinguished from a model that never observed the data to begin with. We develop a certified-removal mechanism for linear classifiers and empirically study learning settings in which this mechanism is practical.

Federated learning was proposed with an intriguing vision of achieving collaborative machine learning among numerous clients without uploading their private data to a cloud server. However, the conventional framework requires each client to leverage the full model for learning, which can be prohibitively inefficient for resource-constrained clients and large-scale deep learning tasks. We thus propose a new framework, called federated submodel learning, where clients download only the needed parts of the full model, namely submodels, and then upload the submodel updates. Nevertheless, the "position" of a client's truly required submodel corresponds to her private data, and its disclosure to the cloud server during interactions inevitably breaks the tenet of federated learning. To integrate efficiency and privacy, we have designed a secure federated submodel learning scheme coupled with a private set union protocol as a cornerstone. Our secure scheme features the properties of randomized response, secure aggregation, and Bloom filter, and endows each client with a customized plausible deniability, in terms of local differential privacy, against the position of her desired submodel, thus protecting her private data. We further instantiated our scheme with the e-commerce recommendation scenario in Alibaba, implemented a prototype system, and extensively evaluated its performance over 30-day Taobao user data. The analysis and evaluation results demonstrate the feasibility and scalability of our scheme from model accuracy and convergency, practical communication, computation, and storage overheads, as well as manifest its remarkable advantages over the conventional federated learning framework.

Despite impressive performance on numerous visual tasks, Convolutional Neural Networks (CNNs) --- unlike brains --- are often highly sensitive to small perturbations of their input, e.g. adversarial noise leading to erroneous decisions. We propose to regularize CNNs using large-scale neuroscience data to learn more robust neural features in terms of representational similarity. We presented natural images to mice and measured the responses of thousands of neurons from cortical visual areas. Next, we denoised the notoriously variable neural activity using strong predictive models trained on this large corpus of responses from the mouse visual system, and calculated the representational similarity for millions of pairs of images from the model's predictions. We then used the neural representation similarity to regularize CNNs trained on image classification by penalizing intermediate representations that deviated from neural ones. This preserved performance of baseline models when classifying images under standard benchmarks, while maintaining substantially higher performance compared to baseline or control models when classifying noisy images. Moreover, the models regularized with cortical representations also improved model robustness in terms of adversarial attacks. This demonstrates that regularizing with neural data can be an effective tool to create an inductive bias towards more robust inference.

Keep it Consistent: T opic-A ware Storytelling from an Image Stream via Iterative Multi-agent Communication Ruize Wang 1, Zhongyu Wei 2, Piji Li 3, Haijun Shan 4, Ji Zhang 4, Qi Zhang 5, Xuanjing Huang 5 1 Academy for Engineering and Technology, Fudan University, China 2 School of Data Science, Fudan University, China 3 Tencent AI Lab, China 4 Zhejiang Lab, China 5 School of Computer Science, Fudan University, China { rzwang18,zywei,qz,xjhuang} @fudan.edu.cn; Abstract Visual storytelling aims to generate a narrative paragraph from a sequence of images automatically. Existing approaches construct text description independently for each image and roughly concatenate them as a story, which leads to the problem of generating semantically incoherent content. In this paper, we proposed a new way for visual storytelling by introducing a topic description task to detect the global semantic context of an image stream. A story is then constructed with the guidance of the topic description. In order to combine the two generation tasks, we propose a multi-agent communication framework that regards the topic description generator and the story generator as two agents and learn them simultaneously via iterative updating mechanism. We validate our approach on VIST, where quantitative results, ablations, and human evaluation demonstrate our method's good ability in generating stories with higher quality compared to state-of-the-art methods. 1 Introduction Image-to-text generation is an important topic in artificial intelligence (AI) which connects computer vision (CV) and natural language processing (NLP). Popular tasks include image captioning (Karpathy and Fei-Fei 2015; Ren et al. 2017; Vinyals et al. 2017) and question answering (Antol et al. 2015; Y u et al. 2017; Fan et al. 2018a; Fan et al. 2018b), aiming at generating a short sentence or a phrase conditioned on certain visual information. It requires the model to understand the main idea of the image stream and generate coherent sentences. Most of existing methods (Huang et al. 2016; Liu et al. 2017; Y u, Bansal, and Berg 2017; Wang et al. 2018a) for visual storytelling extend approaches of image captioning without considering topic information of the image sequence, which causes the problem of generating semantically incoherent content.

Last December, China's top AI scientists gathered in Suzhou for the annual Wu Wenjun AI Science and Technology Award ceremony. They had every reason to expect a feel-good appreciation of China's accomplishments in AI. Yet the mood was decidedly downbeat. "After talking about our advantages, everyone mainly wants to talk about the shortcomings of Chinese AI capabilities in the near-term--where are China's AI weaknesses," said Li Deyi, the president of the Chinese Association for Artificial Intelligence. More than two years after the release of the New Generation Artificial Intelligence Development Plan (AIDP), China's top AI experts worry that Beijing's AI push will not live up to the hype.

With the burning desire and passion to equip ourselves and others with Artificial Intelligence skills, the mission and vision of global AI6 resonated with Azeez Oluwafemi and I the moment it was announced. We were quick to kick-start AI6 in our local community, Lagos, Nigeria on January 6th 2018. We organized and facilitated 16 weeks of study group from January till April and then from August till November. As ambassadors, we played a pivotal role in empowering the AI community in our city to get updated with the latest methods in artificial intelligence, and be able to implement the most cutting-edge AI models out there. For our first cycle, we walked our participants through Computer Vision from Stanford CS231n and Fast AI deep learning for coders.