Scaling up visual category recognition to large numbers of classes remains challenging. A promising research direction is zero-shot learning, which does not require any training data to recognize new classes, but rather relies on some form of auxiliary information describing the new classes. Ultimately, this may allow to use textbook knowledge that humans employ to learn about new classes by transferring knowledge from classes they know well. The most successful zero-shot learning approaches currently require a particular type of auxiliary information -- namely attribute annotations performed by humans -- that is not readily available for most classes. Our goal is to circumvent this bottleneck by substituting such annotations by extracting multiple pieces of information from multiple unstructured text sources readily available on the web. To compensate for the weaker form of auxiliary information, we incorporate stronger supervision in the form of semantic part annotations on the classes from which we transfer knowledge. We achieve our goal by a joint embedding framework that maps multiple text parts as well as multiple semantic parts into a common space. Our results consistently and significantly improve on the state-of-the-art in zero-short recognition and retrieval.
We introduce Markov substitute processes, a new model at the crossroad of statistics and formal grammars, and prove its main property : Markov substitute processes with a given support form an exponential family.
The Support Vector Machine using Privileged Information (SVM+) has been proposed to train a classifier to utilize the additional privileged information that is only available in the training phase but not available in the test phase. In this work, we propose an efficient solution for SVM+ by simply utilizing the squared hinge loss instead of the hinge loss as in the existing SVM+ formulation, which interestingly leads to a dual form with less variables and in the same form with the dual of the standard SVM. The proposed algorithm is utilized to leverage the additional web knowledge that is only available during training for the image categorization tasks. The extensive experimental results on both Caltech101 andWebQueries datasets show that our proposed method can achieve a factor of up to hundred times speedup with the comparable accuracy when compared with the existing SVM+ method.
Data-driven approaches for edge detection have proven effective and achieve top results on modern benchmarks. However, all current data-driven edge detectors require manual supervision for training in the form of hand-labeled region segments or object boundaries. Specifically, human annotators mark semantically meaningful edges which are subsequently used for training. Is this form of strong, high-level supervision actually necessary to learn to accurately detect edges? In this work we present a simple yet effective approach for training edge detectors without human supervision. To this end we utilize motion, and more specifically, the only input to our method is noisy semi-dense matches between frames. We begin with only a rudimentary knowledge of edges (in the form of image gradients), and alternate between improving motion estimation and edge detection in turn. Using a large corpus of video data, we show that edge detectors trained using our unsupervised scheme approach the performance of the same methods trained with full supervision (within 3-5%). Finally, we show that when using a deep network for the edge detector, our approach provides a novel pre-training scheme for object detection.
We propose a method for visual question answering which combines an internal representation of the content of an image with information extracted from a general knowledge base to answer a broad range of image-based questions. This allows more complex questions to be answered using the predominant neural network-based approach than has previously been possible. It particularly allows questions to be asked about the contents of an image, even when the image itself does not contain the whole answer. The method constructs a textual representation of the semantic content of an image, and merges it with textual information sourced from a knowledge base, to develop a deeper understanding of the scene viewed. Priming a recurrent neural network with this combined information, and the submitted question, leads to a very flexible visual question answering approach. We are specifically able to answer questions posed in natural language, that refer to information not contained in the image. We demonstrate the effectiveness of our model on two publicly available datasets, Toronto COCO-QA and MS COCO-VQA and show that it produces the best reported results in both cases.
In this work, we present definition of intuitionistic fuzzy parameterized (IFP) intuitionistic fuzzy soft set and its operations. Then we define IFP-aggregation operator to form IFP-intuitionistic fuzzy soft-decision-making method which allows constructing more efficient decision processes.
Closed Form is a propagation based matting algorithm, functioning well on images with good propagation . The deficiency of the Closed Form method is that for complex areas with poor image propagation , such as hole areas or areas of long and narrow structures. The right results are usually hard to get. On these areas, if certain flags are provided, it can improve the effects of matting. In this paper, we design a matting algorithm by local sampling and the KNN classifier propagation based matting algorithm. First of all, build the corresponding features space according to the different components of image colors to reduce the influence of overlapping between the foreground and background, and to improve the classification accuracy of KNN classifier. Second, adaptively use local sampling or using local KNN classifier for processing based on the pros and cons of the sample performance of unknown image areas. Finally, based on different treatment methods for the unknown areas, we will use different weight for augmenting constraints to make the treatment more effective. In this paper, by combining qualitative observation and quantitative analysis, we will make evaluation of the experimental results through online standard set of evaluation tests. It shows that on images with good propagation , this method is as effective as the Closed Form method, while on images in complex regions, it can perform even better than Closed Form.
Recent applications in computer vision have come to heavily rely on superpixel over-segmentation as a pre-processing step for higher level vision tasks, such as object recognition, image labelling or image segmentation. Here we present a new superpixel algorithm called Hierarchical Piecewise-Constant Super-regions (HPCS), which not only obtains superpixels comparable to the state-of-the-art, but can also be applied hierarchically to form what we call n-th order super-regions. In essence, a Markov Random Field (MRF)-based anisotropic denoising formulation over the quantized feature space is adopted to form piecewise-constant image regions, which are then combined with a graph-based split & merge post-processing step to form superpixels. The graph and quantized feature based formulation of the problem allows us to generalize it hierarchically to preserve boundary adherence with fewer superpixels. Experimental results show that, despite the simplicity of our framework, it is able to provide high quality superpixels, and to hierarchically apply them to form layers of over-segmentation, each with a decreasing number of superpixels, while maintaining the same desired properties (such as adherence to strong image edges). The algorithm is also memory efficient and has a low computational cost.
We introduce a model for constructing vector representations of words by composing characters using bidirectional LSTMs. Relative to traditional word representation models that have independent vectors for each word type, our model requires only a single vector per character type and a fixed set of parameters for the compositional model. Despite the compactness of this model and, more importantly, the arbitrary nature of the form-function relationship in language, our "composed" word representations yield state-of-the-art results in language modeling and part-of-speech tagging. Benefits over traditional baselines are particularly pronounced in morphologically rich languages (e.g., Turkish).
We introduce a class of extensive form games where players might not be able to foresee the possible consequences of their decisions and form a model of their opponents which they exploit to achieve a more profitable outcome. We improve upon existing models of games with limited foresight, endowing players with the ability of higher-order reasoning and proposing a novel solution concept to address intuitions coming from real game play. We analyse the resulting equilibria, devising an effective procedure to compute them.
Morphological reinflection is the task of generating a target form given a source form, a source tag and a target tag. We propose a new way of modeling this task with neural encoder-decoder models. Our approach reduces the amount of required training data for this architecture and achieves state-of-the-art results, making encoder-decoder models applicable to morphological reinflection even for low-resource languages. We further present a new automatic correction method for the outputs based on edit trees.
We present novel methods for analyzing the activation patterns of RNNs from a linguistic point of view and explore the types of linguistic structure they learn. As a case study, we use a multi-task gated recurrent network architecture consisting of two parallel pathways with shared word embeddings trained on predicting the representations of the visual scene corresponding to an input sentence, and predicting the next word in the same sentence. Based on our proposed method to estimate the amount of contribution of individual tokens in the input to the final prediction of the networks we show that the image prediction pathway: a) is sensitive to the information structure of the sentence b) pays selective attention to lexical categories and grammatical functions that carry semantic information c) learns to treat the same input token differently depending on its grammatical functions in the sentence. In contrast the language model is comparatively more sensitive to words with a syntactic function. Furthermore, we propose methods to ex- plore the function of individual hidden units in RNNs and show that the two pathways of the architecture in our case study contain specialized units tuned to patterns informative for the task, some of which can carry activations to later time steps to encode long-term dependencies.
Semantic parsing aims at mapping natural language to machine interpretable meaning representations. Traditional approaches rely on high-quality lexicons, manually-built templates, and linguistic features which are either domain- or representation-specific. In this paper we present a general method based on an attention-enhanced encoder-decoder model. We encode input utterances into vector representations, and generate their logical forms by conditioning the output sequences or trees on the encoding vectors. Experimental results on four datasets show that our approach performs competitively without using hand-engineered features and is easy to adapt across domains and meaning representations.
We present a new trace estimator of the matrix whose explicit form is not given but its matrix multiplication to a vector is available. The form of the estimator is similar to the Hutchison stochastic trace estimator, but instead of the random noise vectors in Hutchison estimator, we use small number of probing vectors determined by machine learning. Evaluation of the quality of estimates and bias correction are discussed. An unbiased estimator is proposed for the calculation of the expectation value of a function of traces. In the numerical experiments with random matrices, it is shown that the precision of trace estimates with $\mathcal{O}(10)$ probing vectors determined by the machine learning is similar to that with $\mathcal{O}(10000)$ random noise vectors.
In this paper we present a novel approach for representing and evolving deformable active contours. The method combines piecewise regular B{\'e}zier models and curve evolution defined by local Free Form Deformation. The contour deformation is locally constrained which allows contour convergence with almost linear complexity while adapting to various shape settings and handling topology changes of the active contour. We demonstrate the effectiveness of the new active contour scheme for visual free space perception and segmentation using omnidirectional images acquired by a robot exploring unknown indoor and outdoor environments. Several experiments validate the approach with comparison to state-of-the art parametric and geometric active contours and provide fast and real-time robot free space segmentation and navigation.
As a contribution to interpretable machine learning research, we develop a novel optimization framework for learning accurate and sparse two-level Boolean rules. We consider rules in both conjunctive normal form (AND-of-ORs) and disjunctive normal form (OR-of-ANDs). A principled objective function is proposed to trade classification accuracy and interpretability, where we use Hamming loss to characterize accuracy and sparsity to characterize interpretability. We propose efficient procedures to optimize these objectives based on linear programming (LP) relaxation, block coordinate descent, and alternating minimization. Experiments show that our new algorithms provide very good tradeoffs between accuracy and interpretability.
We propose a general semi-supervised inference framework focused on the estimation of the population mean. As usual in semi-supervised settings, there exists an unlabeled sample of covariate vectors and a labeled sample consisting of covariate vectors along with real-valued responses ("labels"). Otherwise, the formulation is "assumption-lean" in that no major conditions are imposed on the statistical or functional form of the data. We consider both the ideal semi-supervised setting where infinitely many unlabeled samples are available, as well as the ordinary semi-supervised setting in which only a finite number of unlabeled samples is available. Estimators are proposed along with corresponding confidence intervals for the population mean. Theoretical analysis on both the asymptotic distribution and $\ell_2$-risk for the proposed procedures are given. Surprisingly, the proposed estimators, based on a simple form of the least squares method, outperform the ordinary sample mean. The simple, transparent form of the estimator lends confidence to the perception that its asymptotic improvement over the ordinary sample mean also nearly holds even for moderate size samples. The method is further extended to a nonparametric setting, in which the oracle rate can be achieved asymptotically. The proposed estimators are further illustrated by simulation studies and a real data example involving estimation of the homeless population.
We characterize the class of exchangeable feature allocations assigning probability $V_{n,k}\prod_{l=1}^{k}W_{m_{l}}U_{n-m_{l}}$ to a feature allocation of $n$ individuals, displaying $k$ features with counts $(m_{1},\ldots,m_{k})$ for these features. Each element of this class is parametrized by a countable matrix $V$ and two sequences $U$ and $W$ of non-negative weights. Moreover, a consistency condition is imposed to guarantee that the distribution for feature allocations of $n-1$ individuals is recovered from that of $n$ individuals, when the last individual is integrated out. In Theorem 1.1, we prove that the only members of this class satisfying the consistency condition are mixtures of the Indian Buffet Process over its mass parameter $\gamma$ and mixtures of the Beta--Bernoulli model over its dimensionality parameter $N$. Hence, we provide a characterization of these two models as the only, up to randomization of the parameters, consistent exchangeable feature allocations having the required product form.
We revisit the task of learning a Euclidean metric from data. We approach this problem from first principles and formulate it as a surprisingly simple optimization problem. Indeed, our formulation even admits a closed form solution. This solution possesses several very attractive properties: (i) an innate geometric appeal through the Riemannian geometry of positive definite matrices; (ii) ease of interpretability; and (iii) computational speed several orders of magnitude faster than the widely used LMNN and ITML methods. Furthermore, on standard benchmark datasets, our closed-form solution consistently attains higher classification accuracy.
The Chow form of the essential variety in computer vision is calculated. Our derivation uses secant varieties, Ulrich sheaves and representation theory. Numerical experiments show that our formula can detect noisy point correspondences between two images.