Spatial understanding is a fundamental problem with wide-reaching real-world applications. The representation of spatial knowledge is often modeled with spatial templates, i.e., regions of acceptability of two objects under an explicit spatial relationship (e.g., "on," "below," etc.). In contrast with prior work that restricts spatial templates to explicit spatial prepositions (e.g., "glass on table"), here we extend this concept to implicit spatial language, i.e., those relationships (generally actions) for which the spatial arrangement of the objects is only implicitly implied (e.g., "man riding horse"). In contrast with explicit relationships, predicting spatial arrangements from implicit spatial language requires significant common sense spatial understanding. Here, we introduce the task of predicting spatial templates for two objects under a relationship, which can be seen as a spatial question-answering task with a (2D) continuous output ("where is the man w.r.t. a horse when the man is walking the horse?"). We present two simple neural-based models that leverage annotated images and structured text to learn this task. The good performance of these models reveals that spatial locations are to a large extent predictable from implicit spatial language. Crucially, the models attain similar performance in a challenging generalized setting, where the object-relation-object combinations (e.g., "man walking dog") have never been seen before. Next, we go one step further by presenting the models with unseen objects (e.g., "dog"). In this scenario, we show that leveraging word embeddings enables the models to output accurate spatial predictions, proving that the models acquire solid common sense spatial knowledge allowing for such generalization.

Class imbalance presents a major hurdle in the application of classification methods. A commonly taken approach is to learn ensembles of classifiers using rebalanced data. Examples include bootstrap averaging (bagging) combined with either undersampling or oversampling of the minority class examples. However, rebalancing methods entail asymmetric changes to the examples of different classes, which in turn can introduce their own biases. Furthermore, these methods often require specifying the performance measure of interest a priori, i.e., before learning. An alternative is to employ the threshold moving technique, which applies a threshold to the continuous output of a model, offering the possibility to adapt to a performance measure a posteriori , i.e., a plug-in method. Surprisingly, little attention has been paid to this combination of a bagging ensemble and threshold-moving. In this paper, we study this combination and demonstrate its competitiveness. Contrary to the other resampling methods, we preserve the natural class distribution of the data resulting in well-calibrated posterior probabilities. Additionally, we extend the proposed method to handle multiclass data. We validated our method on binary and multiclass benchmark data sets by using both, decision trees and neural networks as base classifiers. We perform analyses that provide insights into the proposed method.

Spatial understanding is crucial in many real-world problems, yet little progress has been made towards building representations that capture spatial knowledge. Here, we move one step forward in this direction and learn such representations by leveraging a task consisting in predicting continuous 2D spatial arrangements of objects given object-relationship-object instances (e.g., “cat under chair”) and a simple neural network model that learns the task from annotated images. We show that the model succeeds in this task and, furthermore, that it is capable of predicting correct spatial arrangements for unseen objects if either CNN features or word embeddings of the objects are provided. The differences between visual and linguistic features are discussed. Next, to evaluate the spatial representations learned in the previous task, we introduce a task and a dataset consisting in a set of crowdsourced human ratings of spatial similarity for object pairs. We find that both CNN (convolutional neural network) features and word embeddings predict human judgments of similarity well and that these vectors can be further specialized in spatial knowledge if we update them when training the model that predicts spatial arrangements of objects. Overall, this paper paves the way towards building distributed spatial representations, contributing to the understanding of spatial expressions in language.

Language and vision provide complementary information. Integrating both modalities in a single multimodal representation is an unsolved problem with wide-reaching applications to both natural language processing and computer vision. In this paper, we present a simple and effective method that learns a language-to-vision mapping and uses its output visual predictions to build multimodal representations. In this sense, our method provides a cognitively plausible way of building representations, consistent with the inherently re-constructive and associative nature of human memory. Using seven benchmark concept similarity tests we show that the mapped (or imagined) vectors not only help to fuse multimodal information, but also outperform strong unimodal baselines and state-of-the-art multimodal methods, thus exhibiting more human-like judgments. Ultimately, the present work sheds light on fundamental questions of natural language understanding concerning the fusion of vision and language such as the plausibility of more associative and re-constructive approaches.

Human concept representations are often grounded with visual information, yet some aspects of meaning cannot be visually represented or are better described with language. Thus, vision and language provide complementary information that, properly combined, can potentially yield more complete concept representations. Recently, state-of-the-art distributional semantic models and convolutional neural networks have achieved great success in representing linguistic and visual knowledge respectively. In this paper, we compare both, visual and linguistic representations in their ability to capture different types of fine-grain semantic knowledge—or attributes—of concepts. Humans often describe objects using attributes, that is, properties such as shape, color or functionality, which often transcend the linguistic and visual modalities. In our setting, we evaluate how well attributes can be predicted by using the unimodal representations as inputs. We are interested in first, finding out whether attributes are generally better captured by either the vision or by the language modality; and second, if none of them is clearly superior (as we hypothesize), what type of attributes or semantic knowledge are better encoded from each modality. Ultimately, our study sheds light on the potential of combining visual and textual representations.