In recent years there has been a cascade of research in attempting to make AI systems more interpretable by providing explanations; so-called Explainable AI (XAI). Most of this research has dealt with the challenges that arise in explaining black-box deep learning systems in classification and regression tasks, with a focus on tabular and image data; for example, there is a rich seam of work on post-hoc counterfactual explanations for a variety of black-box classifiers (e.g., when a user is refused a loan, the counterfactual explanation tells the user about the conditions under which they would get the loan). However, less attention has been paid to the parallel interpretability challenges arising in AI systems dealing with time series data. This paper advances a novel technique, called Native-Guide, for the generation of proximal and plausible counterfactual explanations for instance-based time series classification tasks (e.g., where users are provided with alternative time series to explain how a classification might change). The Native-Guide method retrieves and uses native in-sample counterfactuals that already exist in the training data as "guides" for perturbation in time series counterfactual generation. This method can be coupled with both Euclidean and Dynamic Time Warping (DTW) distance measures. After illustrating the technique on a case study involving a climate classification task, we reported on a comprehensive series of experiments on both real-world and synthetic data sets from the UCR archive. These experiments provide computational evidence of the quality of the counterfactual explanations generated.

We consider the problem of learning an unknown ReLU network with respect to Gaussian inputs and obtain the first nontrivial results for networks of depth more than two. We give an algorithm whose running time is a fixed polynomial in the ambient dimension and some (exponentially large) function of only the network's parameters. Our bounds depend on the number of hidden units, depth, spectral norm of the weight matrices, and Lipschitz constant of the overall network (we show that some dependence on the Lipschitz constant is necessary). We also give a bound that is doubly exponential in the size of the network but is independent of spectral norm. These results provably cannot be obtained using gradient-based methods and give the first example of a class of efficiently learnable neural networks that gradient descent will fail to learn. In contrast, prior work for learning networks of depth three or higher requires exponential time in the ambient dimension, even when the above parameters are bounded by a constant. Additionally, all prior work for the depth-two case requires well-conditioned weights and/or positive coefficients to obtain efficient run-times. Our algorithm does not require these assumptions. Our main technical tool is a type of filtered PCA that can be used to iteratively recover an approximate basis for the subspace spanned by the hidden units in the first layer. Our analysis leverages new structural results on lattice polynomials from tropical geometry.

There are now many meta-learning methods, each focusing on different modeling aspects of base and meta learners, but all can be (re)formulated as specific bilevel optimization problems. This work presents BOML, a modularized optimization library that unifies several meta-learning algorithms into a common bilevel optimization framework. It provides a hierarchical optimization pipeline together with a variety of iteration modules, which can be used to solve the mainstream categories of meta-learning methods, such as meta-feature-based and meta-initialization-based formulations.

Transfer of pre-trained representations can improve sample efficiency and reduce computational requirements for new tasks. However, representations used for transfer are usually generic, and are not tailored to a particular distribution of downstream tasks. We explore the use of expert representations for transfer with a simple, yet effective, strategy. We train a diverse set of experts by exploiting existing label structures, and use cheap-to-compute performance proxies to select the relevant expert for each target task. This strategy scales the process of transferring to new tasks, since it does not revisit the pre-training data during transfer. Accordingly, it requires little extra compute per target task, and results in a speed-up of 2-3 orders of magnitude compared to competing approaches. Further, we provide an adapter-based architecture able to compress many experts into a single model. We evaluate our approach on two different data sources and demonstrate that it outperforms baselines on over 20 diverse vision tasks in both cases.

Natural language dialogue systems raise great attention recently. As many dialogue models are data-driven, high quality datasets are essential to these systems. In this paper, we introduce Pchatbot, a large scale dialogue dataset which contains two subsets collected from Weibo and Judical forums respectively. Different from existing datasets which only contain post-response pairs, we include anonymized user IDs as well as timestamps. This enables the development of personalized dialogue models which depend on the availability of users' historical conversations. Furthermore, the scale of Pchatbot is significantly larger than existing datasets, which might benefit the data-driven models. Our preliminary experimental study shows that a personalized chatbot model trained on Pchatbot outperforms the corresponding ad-hoc chatbot models. We also demonstrate that using larger dataset improves the quality of dialog models.

Open domain question answering (OpenQA) tasks have been recently attracting more and more attention from the natural language processing (NLP) community. In this work, we present the first free-form multiple-choice OpenQA dataset for solving medical problems, MedQA, collected from the professional medical board exams. It covers three languages: English, simplified Chinese, and traditional Chinese, and contains 12,723, 34,251, and 14,123 questions for the three languages, respectively. We implement both rule-based and popular neural methods by sequentially combining a document retriever and a machine comprehension model. Through experiments, we find that even the current best method can only achieve 36.7\%, 42.0\%, and 70.1\% of test accuracy on the English, traditional Chinese, and simplified Chinese questions, respectively. We expect MedQA to present great challenges to existing OpenQA systems and hope that it can serve as a platform to promote much stronger OpenQA models from the NLP community in the future.

The interactions between the Sun, interplanetary space, near Earth space environment, the Earth's surface, and the power grid are, perhaps unsurprisingly, very complicated. The study of such requires the collaboration between many different organizations spanning the public and private sectors. Thus, an important component of studying space weather is the integration and analysis of heterogeneous information. As such, we have developed a modular ontology to drive the core of the data integration and serve the needs of a highly interdisciplinary community. This paper presents our preliminary modular ontology, for space weather research, as well as demonstrate a method for adaptation to a particular use-case, through the use of existential rules and explicit typing.

Nowcasting is a field of meteorology which aims at forecasting weather on a short term of up to a few hours. In the meteorology landscape, this field is rather specific as it requires particular techniques, such as data extrapolation, where conventional meteorology is generally based on physical modeling. In this paper, we focus on cloud cover nowcasting, which has various application areas such as satellite shots optimisation and photovoltaic energy production forecast. Following recent deep learning successes on multiple imagery tasks, we applied deep convolutionnal neural networks on Meteosat satellite images for cloud cover nowcasting. We present the results of several architectures specialized in image segmentation and time series prediction. We selected the best models according to machine learning metrics as well as meteorological metrics. All selected architectures showed significant improvements over persistence and the well-known U-Net surpasses AROME physical model.

The next year will be pivotal for the Air Force's effort to acquire a new class of autonomous drones, as industry teams compete for a chance to build a fleet of robotic wingmen that will soon undergo operational experimentation. The "Skyborg" program is one of the service's top science-and-technology priorities under the "Vanguard" initiative to deliver game-changing capabilities to its warfighters. The aim is to acquire relatively inexpensive, attritable unmanned aircraft that can leverage artificial intelligence and accompany manned fighter jets into battle. "I expect that we will do sorties where a set number are expected to fly with the manned systems, and we'll have crazy new [concepts of operation] for how they'll be used," Assistant Secretary of the Air Force for Acquisition, Technology and Logistics Will Roper said during an online event hosted by the Mitchell Institute for Aerospace Studies. The platforms might even be called upon to conduct kamikaze missions.

We address the question of characterizing and finding optimal representations for supervised learning. Traditionally, this question has been tackled using the Information Bottleneck, which compresses the inputs while retaining information about the targets, in a decoder-agnostic fashion. In machine learning, however, our goal is not compression but rather generalization, which is intimately linked to the predictive family or decoder of interest (e.g. linear classifier). We propose the Decodable Information Bottleneck (DIB) that considers information retention and compression from the perspective of the desired predictive family. As a result, DIB gives rise to representations that are optimal in terms of expected test performance and can be estimated with guarantees. Empirically, we show that the framework can be used to enforce a small generalization gap on downstream classifiers and to predict the generalization ability of neural networks.