Task-oriented semantic parsing is a critical component of virtual assistants, which is responsible for understanding the user's intents (set reminder, play music, etc.). Recent advances in deep learning have enabled several approaches to successfully parse more complex queries Rongali et al., 2020), but these models require a large amount of annotated training data to parse queries on new domains (e.g. reminder, music).In this paper, we focus on adapting taskoriented semantic parsers to low-resource domains, and propose a novel method that outperforms a supervised neural model at a 10-fold data reduction. In particular, we identify two fundamental factors for low-resource domain adaptation: better representation learning and better training techniques. Our representation learning uses BART (Lewis et al., 2020) to initialize our model which outperforms encoder-only pre-trained representations used in previous work. Furthermore, we train with optimization-based meta-learning (Finn et al., 2017) to improve generalization to lowresource domains. This approach significantly outperforms all baseline methods in the experiments on a newly collected multi-domain taskoriented semantic parsing dataset (TOPv2 1 ).

BackgroundMicroRNAs (miRNAs) are small regulatory RNA that mediate RNA interference by binding to various mRNA target regions. There have been several computational methods for the identification of target mRNAs for miRNAs. However, these have considered all contributory features as scalar representations, primarily, as thermodynamic or sequence-based features. Further, a majority of these methods solely target canonical sites, which are sites with “seed” complementarity. Here, we present a machine-learning classification scheme, titled Avishkar, which captures the spatial profile of miRNA-mRNA interactions via smooth B-spline curves, separately for various input features, such as thermodynamic and sequence features. Further, we use a principled approach to uniformly model canonical and non-canonical seed matches, using a novel seed enrichment metric.ResultsWe demonstrate that large number of seed-match patterns have high enrichment values, conserved across species, and that majority of miRNA binding sites involve non-canonical matches, corroborating recent findings. Using spatial curves and popular categorical features, such as target site length and location, we train a linear SVM model, utilizing experimental CLIP-seq data. Our model significantly outperforms all established methods, for both canonical and non-canonical sites. We achieve this while using a much larger candidate miRNA-mRNA interaction set than prior work.ConclusionsWe have developed an efficient SVM-based model for miRNA target prediction using recent CLIP-seq data, demonstrating superior performance, evaluated using ROC curves, specifically about 20 % better than the state-of-the-art, for different species (human or mouse), or different target types (canonical or non-canonical). To the best of our knowledge we provide the first distributed framework for microRNA target prediction based on Apache Hadoop and Spark.AvailabilityAll source code and data is publicly available at https://bitbucket.org/cellsandmachines/avishkar.

In this paper we study the problem of exact recovery of the pure-strategy Nash equilibria (PSNE) set of a graphical game from noisy observations of joint actions of the players alone. We consider sparse linear influence games -a parametric class of graphical games with linear payoffs, and represented by directed graphs of n nodes (players) and in-degree of at most k. We present an 1 -regularized logistic regression based algorithm for recovering the PSNE set exactly, that is both computationally efficient -i.e. runs in polynomial time -and statistically efficient -i.e. has logarithmic sample complexity. Specifically, we show that the sufficient number of samples required for exact PSNE recovery scales as O (poly(k) log n). We also validate our theoretical results using synthetic experiments.

Natural language (NL) explanations of model predictions are gaining popularity as a means to understand and verify decisions made by large black-box pre-trained models, for tasks such as Question Answering (QA) and Fact Verification. Recently, pre-trained sequence to sequence (seq2seq) models have proven to be very effective in jointly making predictions, as well as generating NL explanations. However, these models have many shortcomings; they can fabricate explanations even for incorrect predictions, they are difficult to adapt to long input documents, and their training requires a large amount of labeled data. In this paper, we develop FiD-Ex 1 , which addresses these shortcomings for seq2seq models by: 1) introducing sentence markers to eliminate explanation fabrication by encouraging extractive generation, 2) using the fusion-in-decoder architecture to handle long input contexts, and 3) intermediate fine-tuning on re-structured open domain QA datasets to improve few-shot performance. FiD-Ex significantly improves over prior work in terms of explanation metrics and task accuracy on five tasks from the ERASER explainability benchmark in both fully supervised and few-shot settings.

Natural language (NL) explanations of model predictions are gaining popularity as a means to understand and verify decisions made by large black-box pre-trained models, for NLP tasks such as Question Answering (QA) and Fact Verification. Recently, pre-trained sequence to sequence (seq2seq) models have proven to be very effective in jointly making predictions, as well as generating NL explanations. However, these models have many shortcomings; they can fabricate explanations even for incorrect predictions, they are difficult to adapt to long input documents, and their training requires a large amount of labeled data. In this paper, we develop FiD-Ex, which addresses these shortcomings for seq2seq models by: 1) introducing sentence markers to eliminate explanation fabrication by encouraging extractive generation, 2) using the fusionin-decoder architecture to handle long input contexts, and 3) intermediate fine-tuning on re-structured open domain QA datasets to improve few-shot performance. FiD-Ex significantly improves over prior work in terms of explanation metrics and task accuracy, on multiple tasks from the ERASER explainability benchmark, both in the fully supervised and in the few-shot settings. 1 In this work, we use textual rationales and NL explanations interchangeably