Our goal is to answer real-world tourism questions that seek Points-of-Interest (POI) recommendations. Such questions express various kinds of spatial and non-spatial constraints, necessitating a combination of textual and spatial reasoning. In response, we develop the first joint spatio-textual reasoning model, which combines geo-spatial knowledge with information in textual corpora to answer questions. We first develop a modular spatial-reasoning network that uses geo-coordinates of location names mentioned in a question, and of candidate answer POIs, to reason over only spatial constraints. We then combine our spatial-reasoner with a textual reasoner in a joint model and present experiments on a real world POI recommendation task. We report substantial improvements over existing models with-out joint spatio-textual reasoning.
Recent research has proposed neural architectures for solving combinatorial problems in structured output spaces. In many such problems, there may exist multiple solutions for a given input, e.g. a partially filled Sudoku puzzle may have many completions satisfying all constraints. Further, we are often interested in finding {\em any one} of the possible solutions, without any preference between them. Existing approaches completely ignore this solution multiplicity. In this paper, we argue that being oblivious to the presence of multiple solutions can severely hamper their training ability. Our contribution is two fold. First, we formally define the task of learning one-of-many solutions for combinatorial problems in structured output spaces, which is applicable for solving several problems of interest such as N-Queens, and Sudoku. Second, we present a generic learning framework that adapts an existing prediction network for a combinatorial problem to handle solution multiplicity. Our framework uses a selection module, whose goal is to dynamically determine, for every input, the solution that is most effective for training the network parameters in any given learning iteration. We propose an RL based approach to jointly train the selection module with the prediction network. Experiments on three different domains, and using two different prediction networks, demonstrate that our framework significantly improves the accuracy in our setting, obtaining up to $21$ pt gain over the baselines.
A robot working in a physical environment (like home or factory) needs to learn to use various available tools for accomplishing different tasks, for instance, a mop for cleaning and a tray for carrying objects. The number of possible tools is large and it may not be feasible to demonstrate usage of each individual tool during training. Can a robot learn commonsense knowledge and adapt to novel settings where some known tools are missing, but alternative unseen tools are present? We present a neural model that predicts the best tool from the available objects for achieving a given declarative goal. This model is trained by user demonstrations, which we crowd-source through humans instructing a robot in a physics simulator. This dataset maintains user plans involving multi-step object interactions along with symbolic state changes. Our neural model, ToolNet, combines a graph neural network to encode the current environment state, and goal-conditioned spatial attention to predict the appropriate tool. We find that providing metric and semantic properties of objects, and pre-trained object embeddings derived from a commonsense knowledge repository such as ConceptNet, significantly improves the model's ability to generalize to unseen tools. The model makes accurate and generalizable tool predictions. When compared to a graph neural network baseline, it achieves 14-27% accuracy improvement for predicting known tools from new world scenes, and 44-67% improvement in generalization for novel objects not encountered during training.
While traditional systems for Open Information Extraction were statistical and rule-based, recently neural models have been introduced for the task. Our work builds upon CopyAttention, a sequence generation OpenIE model (Cui et. al., 2018). Our analysis reveals that CopyAttention produces a constant number of extractions per sentence, and its extracted tuples often express redundant information. We present IMoJIE, an extension to CopyAttention, which produces the next extraction conditioned on all previously extracted tuples. This approach overcomes both shortcomings of CopyAttention, resulting in a variable number of diverse extractions per sentence. We train IMoJIE on training data bootstrapped from extractions of several non-neural systems, which have been automatically filtered to reduce redundancy and noise. IMoJIE outperforms CopyAttention by about 18 F1 pts, and a BERT-based strong baseline by 2 F1 pts, establishing a new state of the art for the task.
Knowledge Base Completion has been a very active area recently, where multiplicative models have generally outperformed additive and other deep learning methods -- like GNN, CNN, path-based models. Several recent KBC papers propose architectural changes, new training methods, or even a new problem reformulation. They evaluate their methods on standard benchmark datasets - FB15k, FB15k-237, WN18, WN18RR, and Yago3-10. Recently, some papers discussed how 1-N scoring can speed up training and evaluation. In this paper, we discuss how by just applying this training regime to a basic model like Complex gives near SOTA performance on all the datasets -- we call this model COMPLEX-V2. We also highlight how various multiplicative methods recently proposed in literature benefit from this trick and become indistinguishable in terms of performance on most datasets. This paper calls for a reassessment of their individual value, in light of these findings.
Temporal knowledge bases associate relational (s,r,o) triples with a set of times (or a single time instant) when the relation is valid. While time-agnostic KB completion (KBC) has witnessed significant research, temporal KB completion (TKBC) is in its early days. In this paper, we consider predicting missing entities (link prediction) and missing time intervals (time prediction) as joint TKBC tasks where entities, relations, and time are all embedded in a uniform, compatible space. We present TIMEPLEX, a novel time-aware KBC method, that also automatically exploits the recurrent nature of some relations and temporal interactions between pairs of relations. TIMEPLEX achieves state-of-the-art performance on both prediction tasks. We also find that existing TKBC models heavily overestimate link prediction performance due to imperfect evaluation mechanisms. In response, we propose improved TKBC evaluation protocols for both link and time prediction tasks, dealing with subtle issues that arise from the partial overlap of time intervals in gold instances and system predictions.
Two common types of tasks on Knowledge Bases have been studied -- single link prediction (Knowledge Base Completion) and path query answering. However, our analysis of user queries on a real-world knowledge base reveals that a significant fraction of queries specify paths using regular expressions(regex). Such regex queries cannot be handled by any of the existing link prediction or path query answering models. In response, we present Regex Query Answering, the novel task of answering regex queries on incomplete KBs. We contribute two datasets for the task, including one where test queries are harvested from actual user querylogs. We train baseline neural models for our new task and propose novel ways to handle disjunction and Kleene plus regex operators.
Pooling-based recurrent neural architectures consistently outperform their counterparts without pooling. However, the reasons for their enhanced performance are largely unexamined. In this work, we examine three commonly used pooling techniques (mean-pooling, max-pooling, and attention), and propose max-attention, a novel variant that effectively captures interactions among predictive tokens in a sentence. We find that pooling-based architectures substantially differ from their non-pooling equivalents in their learning ability and positional biases--which elucidate their performance benefits. By analyzing the gradient propagation, we discover that pooling facilitates better gradient flow compared to BiLSTMs. Further, we expose how BiLSTMs are positionally biased towards tokens in the beginning and the end of a sequence. Pooling alleviates such biases. Consequently, we identify settings where pooling offers large benefits: (i) in low resource scenarios, and (ii) when important words lie towards the middle of the sentence. Among the pooling techniques studied, max-attention is the most effective, resulting in significant performance gains on several text classification tasks.
Task-oriented dialog (TOD) systems converse with users to accomplish a specific task. This task requires the system to query a knowledge base (KB) and use the retrieved results to fulfil user needs. Predicting the KB queries is crucial and can lead to severe under-performance if made incorrectly. KB queries are usually annotated in real-world datasets and are learnt using supervised approaches to achieve acceptable task completion. This need for query annotations prevents TOD systems from easily adapting to new domains. In this paper, we propose a novel problem of learning end-to-end TOD systems using dialogs that do not contain KB query annotations. Our approach first learns to predict the KB queries using reinforcement learning (RL) and then learns the end-to-end system using the predicted queries. However, predicting the correct query in TOD systems is uniquely plagued by correlated attributes, in which, due to data bias, certain attributes always occur together in the KB. This prevents the RL system to generalise and accuracy suffers as a result. We propose Correlated Attributes Resilient RL (CARRL), a modification to the RL gradient estimation, which mitigates the problem of correlated attributes and predicts KB queries better than existing weakly supervised approaches. Finally, we compare the performance of our end-to-end system trained using predicted queries to a system trained using annotated gold queries.