United States

Recent developments in pre-trained neural language modeling have led to leaps in accuracy on common-sense question-answering benchmarks. However,  there is increasing concern that models overfit to specific tasks, without learning to utilize external knowledge or perform general semantic reasoning. In contrast, zero-shot evaluations have shown promise as a more robust measure of a modelâ€™s general reasoning abilities. In this paper, we propose a novel neuro-symbolic framework for zero-shot question answering across commonsense tasks. Guided by a set of hypotheses,  the framework studies how to transform various pre-existing knowledge resources into a form that is most effective for pre-training models. We vary the set of language models, training regimes, knowledge sources, and data generation strategies, and measure their impact across tasks. Extending on prior work, we devise and compare four constrained distractor-sampling strategies. We provide empirical results across five commonsense question-answering tasks with data generated from five external knowledge resources. We show that, while an individual knowledge graph is better suited for specific tasks, a global knowledge graph brings consistent gains across different tasks. In addition, both preserving the structure of the task as well as generating fair and informative questions help language models learn more effectively.

AAAI 2021

Knowledge-Driven Data Construction for Zero-Shot Evaluation in Commonsense Question Answering

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

technical paper

Non-negative matrix factorization (NMF) is a highly celebrated algorithm for matrix decomposition that guarantees non-negative factors. The underlying optimization problem is computationally intractable, yet in practice, gradient-descent-based methods often find good solutions. In this paper, we revisit the NMF optimization problem and analyze its loss landscape in non-worst-case settings. It has recently been observed that gradients in deep networks tend to point towards the final minimizer throughout the optimization procedure. We show that a similar property holds (with high probability) for NMF, provably in a non-worst case model with a planted solution, and empirically across an extensive suite of real-world NMF problems. Our analysis predicts that this property becomes more likely with growing number of parameters, and experiments suggest that a similar trend might also hold for deep neural networks---turning increasing dataset sizes and model sizes into a blessing from an optimization perspective.

Characterizing the Loss Landscape in Non-Negative Matrix Factorization

We propose a novel high-performance and interpretable canonical deep tabular data learning architecture, TabNet. TabNet uses sequential attention to choose which features to reason from at each decision step, enabling interpretability and more efficient learning as the learning capacity is used for the most salient features. We demonstrate that TabNet outperforms other variants on a wide range of non-performance-saturated tabular datasets and yields interpretable feature attributions plus insights into its global behavior. Finally, we demonstrate self-supervised learning for tabular data, significantly improving performance when unlabeled data is abundant.

TabNet: Attentive Interpretable Tabular Learning

In this paper, we study the computational complexity of the Maximum Satisfiability problem in terms of the length L of a given formula. We present an algorithm with running time O(1.0927^L), hence, improving the previously known best upper bound O(1.1058^L) developed more than 20 years ago by Bansal and Raman.  Theoretically speaking, our algorithm increases the length of solvable formulas by 13.3% (compare this to the recent breakthrough result for Maximum Satisfiability problem with respect to the number of clauses by Xu et al. in 2019 giving a 7.5% improvement). Besides, we propose a significantly simpler algorithm with running time O(1.1049^L). The algorithm outperforms Bansal's and Raman's algorithm in simplicity and running time.

New Length Dependent Algorithm for Maximum Satisfiability Problem

Neural sequence-to-sequence models are finding increasing use in editing of documents, for example in correcting a text document or repairing source code. In this paper, we argue that common seq2seq models (with a facility to copy single tokens) are not a natural fit for such tasks, as they have to explicitly copy each unchanged token. We present an extension of seq2seq models capable of copying entire spans of the input to the output in one step, greatly reducing the number of decisions required during inference. This extension means that there are now many ways of generating the same output, which we handle by deriving a new objective for training and a variation of beam search for inference that explicitly handles this problem. In our experiments on a range of editing tasks of natural language and source code, we show that our new model consistently outperforms simpler baselines.

Copy That! Editing Sequences by Copying Spans

Forecasting on sparse multivariate time series (MTS) aims to model the predictors of future values of time series given their incomplete past, which is important for many emerging applications. However, most existing methods process MTSâ€™s individually, and do not leverage the dynamic distributions underlying the MTSâ€™s, leading to sub-optimal results when the sparsity is high. To address this challenge, we propose a novel generative model, which tracks the transition of latent clusters, instead of isolated feature representations, to achieve robust modeling. It is characterized by a newly designed dynamic Gaussian mixture distribution, which captures the dynamics of clustering structures, and is used for emitting time series. The generative model is parameterized by neural networks. A structured inference network is also designed for enabling inductive analysis.  A gating mechanism is further introduced to dynamically tune the Gaussian mixture distributions. Extensive experimental results on a variety of real-life datasets demonstrate the effectiveness of our method.

Dynamic Gaussian Mixture Based Deep Generative Model for Robust Forecasting on Sparse Multivariate Time Series

Recent advances in unsupervised domain adaptation (UDA) show that transferable prototypical learning presents a powerful means for class conditional alignment, which encourages the closeness of cross-domain class centroids. However, the cross-domain inner-class compactness and the underlying fine-grained subtype structure remained largely underexplored. In this work, we propose to adaptively carry out the fine-grained subtype-aware alignment by explicitly enforcing the class-wise separation and subtype-wise compactness with intermediate pseudo labels. Our key insight is that the unlabeled subtypes of a class can be divergent to one another with different conditional and label shifts, while inheriting the local proximity within a subtype. The cases with or without the prior information on subtype numbers are investigated to discover the underlying subtype structure in an online fashion. The proposed subtype-aware dynamic UDA achieves promising results on a medical diagnosis task.

Subtype-Aware Unsupervised Domain Adaptation for Medical Diagnosis

Schelling's model is an influential model that reveals how individual perceptions and incentives can lead to racial segregation. Inspired by a recent stream of work, we study welfare guarantees and complexity in this model with respect to several welfare measures. First, we show that while maximizing the social welfare is NP-hard, computing an assignment with approximately half of the maximum welfare can be done in polynomial time. We then consider Pareto optimality and introduce two new optimality notions, and establish mostly tight bounds on the worst-case welfare loss for assignments satisfying these notions. In addition, we show that for trees, it is possible to decide whether there exists an assignment that gives every agent a positive utility in polynomial time; moreover, when every node in the topology has degree at least 2, such an assignment always exists and can be found efficiently.

Welfare Guarantees in Schelling Segregation

AI Safety is a major concern in many deep learning applications such as autonomous driving. Given a trained deep learning model, an important natural problem is how to reliably verify the model's prediction. In this paper, we propose a novel framework --- deep verifier networks (DVN) to detect unreliable inputs or predictions of deep discriminative models, using separately trained deep generative models. Our proposed model is based on conditional variational auto-encoders with disentanglement constraints to separate the label information from the latent representation. We give both intuitive and theoretical justifications for the model. Our verifier network is trained independently with the prediction model, which eliminates the need of retraining the verifier network for a new model. We test the verifier network on both out-of-distribution detection and adversarial example detection problems, as well as anomaly detection problems in structured prediction tasks such as image caption generation. We achieve state-of-the-art results in all of these problems.

Deep Verifier Networks: Verification of Deep Discriminative Models with Deep Generative Models

Deep neural networks have been successfully deployed in various domains of artificial intelligence, including computer vision and natural language processing. We observe that the current standard procedure for training DNNs discards all the learned information in the past epochs except the current learned weights. An interesting question is: is this discarded information indeed useless? We argue that the discarded information can benefit the subsequent training. In this paper, we propose learning with retrospection (LWR) which makes use of the learned information in the past epochs to guide the subsequent training. LWR is a simple yet effective training framework to improve accuracies, calibration, and robustness of DNNs without introducing any additional network parameters or inference cost, but only with a negligible training overhead. Extensive experiments on several benchmark datasets demonstrate the superiority of LWR for training DNNs.

Learning with Retrospection

Mobile devices are becoming an important carrier for deep learning tasks, as they are being equipped with powerful, high-end mobile CPUs and GPUs. However, it is still a challenging task to execute 3D Convolutional Neural Networks (CNNs) targeting for real-time performance, besides high inference accuracy. The reason is more complex model structure and higher model dimensionality overwhelm the available computation/storage resources on mobile devices. A natural way may be turning to deep learning weight pruning techniques. However, the direct generalization of existing 2D CNN weight pruning methods to 3D CNNs is not ideal for fully exploiting mobile parallelism while achieving high inference accuracy.

This paper proposes RT3D, a model compression and mobile acceleration framework for 3D CNNs, seamlessly integrating neural network weight pruning and compiler code generation techniques. We propose and investigate two structured sparsity schemes i.e., the vanilla structured sparsity and kernel group structured (KGS) sparsity that are mobile acceleration friendly. The vanilla sparsity removes whole kernel groups, while KGS sparsity is a more fine-grained structured sparsity that enjoys higher flexibility while exploiting full on-device parallelism. We propose a reweighted regularization pruning algorithm to achieve the proposed sparsity schemes. The inference time speedup due to sparsity is approaching the pruning rate of the whole model FLOPs (floating point operations). RT3D demonstrates up to 29.1x speedup in end-to-end inference time comparing with current mobile frameworks supporting 3D CNNs, with moderate 1%~1.5% accuracy loss. The end-to-end inference time for 16 video frames could be within 150 ms, when executing representative C3D and R(2+1)D models on a cellphone. For the first time, real-time execution of 3D CNNs is achieved on off-the-shelf mobiles.

Downloads

Next from AAAI 2021

Characterizing the Loss Landscape in Non-Negative Matrix Factorization

Similar lecture

Dynamic Neuro-Symbolic Knowledge Graph Construction for Zero-Shot Commonsense Question Answering

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from AAAI 2021

Characterizing the Loss Landscape in Non-Negative Matrix Factorization

Similar lecture

Dynamic Neuro-Symbolic Knowledge Graph Construction for Zero-Shot Commonsense Question Answering

Downloads