Neural Module Networks, originally proposed for the task of visual question answering, are a class of neural network architectures that involve human-specified neural modules, each designed for a specific form of reasoning. In current formulations of such networks only the parameters of the neural modules and/or the order of their execution is learned. In this work, we further expand this approach and also learn the underlying internal structure of modules in terms of the ordering and combination of simple and elementary arithmetic operators. We utilize a minimum amount of prior knowledge from the human-specified neural modules in the form of different input types and arithmetic operators used in these modules. Our results show that one is indeed able to simultaneously learn both internal module structure and module sequencing without extra supervisory signals for module execution sequencing. With this approach, we report performance comparable to models using hand-designed modules. In addition, we do a analysis of sensitivity of the learned modules w.r.t. the arithmetic operations and infer the analytical expressions of the learned modules.

In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages. We combine two existing objective functions to make images and captions close in a joint embedding space while adapting the alignment of word embeddings between existing languages in our model. We show that our approach enables better generalization, achieving state-of-the-art performance in text-to-image and image-to-text retrieval task, and caption-caption similarity task. Two multimodal multilingual datasets are used for evaluation: Multi30k with German and English captions and Microsoft-COCO with English and Japanese captions.

In this paper, we experiment with a recently proposed visual reasoning task dealing with quantities – modeling the multimodal, contextually-dependent meaning of size adjectives (‘big’, ‘small’) – and explore the impact of varying the training data on the learning behavior of a state-of-art system. In previous work, models have been shown to fail in generalizing to unseen adjective-noun combinations. Here, we investigate whether, and to what extent, seeing some of these cases during training helps a model understand the rule subtending the task, i.e., that being big implies being not small, and vice versa. We show that relatively few examples are enough to understand this relationship, and that developing a specific, mutually exclusive representation of size adjectives is beneficial to the task.

Chinese characters are unique in its logographic nature, which inherently encodes world knowledge through thousands of years evolution. This paper proposes an embedding approach, namely eigencharacter (EC) space, which helps NLP application easily access the knowledge encoded in Chinese orthography. These EC representations are automatically extracted, encode both structural and radical information, and easily integrate with other computational models. We built EC representations of 5,000 Chinese characters, investigated orthography knowledge encoded in ECs, and demonstrated how these ECs identified visually similar characters with both structural and radical information.

Scene graphs represent semantic information in images, which can help image captioning system to produce more descriptive outputs versus using only the image as context. Recent captioning approaches rely on ad-hoc approaches to obtain graphs for images. However, those graphs introduce noise and it is unclear the effect of parser errors on captioning accuracy. In this work, we investigate to what extent scene graphs can help image captioning. Our results show that a state-of-the-art scene graph parser can boost performance almost as much as the ground truth graphs, showing that the bottleneck currently resides more on the captioning models than on the performance of the scene graph parser.

It is assumed that multimodal machine translation systems are better than text-only systems at translating phrases that have a direct correspondence in the image. This assumption has been challenged in experiments demonstrating that state-of-the-art multimodal systems perform equally well in the presence of randomly selected images, but, more recently, it has been shown that masking entities from the source language sentence during training can help to overcome this problem. In this paper, we conduct experiments with both visual and textual adversaries in order to understand the role of incorrect textual inputs to such systems. Our results show that when the source language sentence contains mistakes, multimodal translation systems do not leverage the additional visual signal to produce the correct translation. We also find that the degradation of translation performance caused by textual adversaries is significantly higher than by visual adversaries.

Previous research into agent communication has shown that a pre-trained guide can speed up the learning process of an imitation learning agent. The guide achieves this by providing the agent with discrete messages in an emerged language about how to solve the task. We extend this one-directional communication by a one-bit communication channel from the learner back to the guide: It is able to ask the guide for help, and we limit the guidance by penalizing the learner for these requests. During training, the agent learns to control this gate based on its current observation. We find that the amount of requested guidance decreases over time and guidance is requested in situations of high uncertainty. We investigate the agent’s performance in cases of open and closed gates and discuss potential motives for the observed gating behavior.

Readers’ eye movements used as part of the training signal have been shown to improve performance in a wide range of Natural Language Processing (NLP) tasks. Previous work uses gaze data either at the type level or at the token level and mostly from a single eye-tracking corpus. In this paper, we analyze type vs token-level integration options with eye tracking data from two corpora to inform two syntactic sequence labeling problems: binary phrase chunking and part-of-speech tagging. We show that using globally-aggregated measures that capture the central tendency or variability of gaze data is more beneficial than proposed local views which retain individual participant information. While gaze data is informative for supervised POS tagging, which complements previous findings on unsupervised POS induction, almost no improvement is obtained for binary phrase chunking, except for a single specific setup. Hence, caution is warranted when using gaze data as signal for NLP, as no single view is robust over tasks, modeling choice and gaze corpus.

How can we teach artificial agents to use human language flexibly to solve problems in real-world environments? We have an example of this in nature: human babies eventually learn to use human language to solve problems, and they are taught with an adult human-in-the-loop. Unfortunately, current machine learning methods (e.g. from deep reinforcement learning) are too data inefficient to learn language in this way. An outstanding goal is finding an algorithm with a suitable ‘language learning prior’ that allows it to learn human language, while minimizing the number of on-policy human interactions. In this paper, we propose to learn such a prior in simulation using an approach we call, Learning to Learn to Communicate (L2C). Specifically, in L2C we train a meta-learning agent in simulation to interact with populations of pre-trained agents, each with their own distinct communication protocol. Once the meta-learning agent is able to quickly adapt to each population of agents, it can be deployed in new populations, including populations speaking human language. Our key insight is that such populations can be obtained via self-play, after pre-training agents with imitation learning on a small amount of off-policy human language data. We call this latter technique Seeded Self-Play (S2P). Our preliminary experiments show that agents trained with L2C and S2P need fewer on-policy samples to learn a compositional language in a Lewis signaling game.