Despite a multitude of empirical studies, little consensus exists on whether neural networks are able to generalise compositionally, a controversy that, in part, stems from a lack of agreement about what it means for a neural model to be compositional. As a response to this controversy, we present a set of tests that provide a bridge between, on the one hand, the vast amount of linguistic and philosophical theory about compositionality and, on the other, the successful neural models of language. We collect different interpretations of compositionality and translate them into five theoretically grounded tests that are formulated on a task-independent level. In particular, we provide tests to investigate (i) if models systematically recombine known parts and rules (ii) if models can extend their predictions beyond the length they have seen in the training data (iii) if models' composition operations are local or global (iv) if models' predictions are robust to synonym substitutions and (v) if models favour rules or exceptions during training. To demonstrate the usefulness of this evaluation paradigm, we instantiate these five tests on a highly compositional data set which we dub PCFG SET and apply the resulting tests to three popular sequence-to- sequence models: a recurrent, a convolution based and a transformer model. We provide an in depth analysis of the results, that uncover the strengths and weaknesses of these three architectures and point to potential areas of improvement.

We consider first order stochastic optimization where the oracle must quantize each subgradient estimate to $r$ bits. We treat two oracle models: the first where the Euclidean norm of the oracle output is almost surely bounded and the second where it is mean square bounded. Prior work in this setting assumes the availability of unbiased quantizers. While this assumption is valid in the case of almost surely bounded oracles, it does not hold true for the standard setting of mean square bounded oracles, and the bias can dramatically affect the convergence rate. We analyze the performance of standard quantizers from prior work in combination with projected stochastic gradient descent for both these oracle models and present two new adaptive quantizers that outperform the existing ones. Specifically, for almost surely bounded oracles, we establish first a lower bound for the precision needed to attain the standard convergence rate of $T^{-\frac 12}$ for optimizing convex functions over a $d$-dimentional domain. Our proposed Rotated Adaptive Tetra-iterated Quantizer (RATQ) is merely a factor $O(\log \log \log^\ast d)$ far from this lower bound. For mean square bounded oracles, we show that a state-of-the-art Rotated Uniform Quantizer (RUQ) from prior work would need atleast $\Omega(d\log T)$ bits to achieve the convergence rate of $T^{-\frac 12}$, using any optimization protocol. However, our proposed Rotated Adaptive Quantizer (RAQ) outperforms RUQ in this setting and attains a convergence rate of $T^{-\frac 12}$ using a precision of only $O(d\log\log T)$. For mean square bounded oracles, in the communication-starved regime where the precision $r$ is fixed to a constant independent of $T$, we show that RUQ cannot attain a convergence rate better than $T^{-\frac 14}$ for any $r$, while RAQ can attain convergence at rates arbitrarily close to $T^{-\frac 12}$ as $r$ increases.

Practical application of Reinforcement Learning (RL) often involves risk considerations. We study a generalized approximation scheme for risk measures, based on Monte-Carlo simulations, where the risk measures need not necessarily be \emph{coherent}. We demonstrate that, even in simple problems, measures such as the variance of the reward-to-go do not capture the risk in a satisfactory manner. In addition, we show how a risk measure can be derived from model's realizations. We propose a neural architecture for estimating the risk and suggest the risk critic architecture that can be use to optimize a policy under general risk measures. We conclude our work with experiments that demonstrate the efficacy of our approach.

The battery is a key component of autonomous robots. Its performance limits the robot's safety and reliability. Unlike liquid-fuel, a battery, as a chemical device, exhibits complicated features, including (i) capacity fade over successive recharges and (ii) increasing discharge rate as the state of charge (SOC) goes down for a given power demand. Existing formal verification studies of autonomous robots, when considering energy constraints, formalise the energy component in a generic manner such that the battery features are overlooked. In this paper, we model an unmanned aerial vehicle (UA V) inspection mission on a wind farm and via probabilistic model checking in PRISM show (i) how the battery features may affect the verification results significantly in practical cases; and (ii) how the battery features, together with dynamic environments and battery safety strategies, jointly affect the verification results. Potential solutions to explicitly integrate battery prognostics and health management (PHM) with formal verification of autonomous robots are also discussed to motivate future work. Keywords: Formal verification · Probabilistic model checking · PRISM · Autonomous systems · Unmanned aerial vehicle · Battery PHM. 1 Introduction Autonomous robots, such as unmanned aerial vehicles (UA V) (commonly termed drones 3), unmanned underwater vehicles (UUV), self-driving cars and legged-robots, obtain increasingly widespread applications in many domains [14].

Mutually beneficial behavior in repeated games can be enforced via the threat of punishment, as enshrined in game theory's well-known "folk theorem." There is a cost, however, to a player for generating these disincentives. In this work, we seek to minimize this cost by computing a "Stackelberg punishment," in which the player selects a behavior that sufficiently punishes the other player while maximizing its own score under the assumption that the other player will adopt a best response. This idea generalizes the concept of a Stackelberg equilibrium. Known efficient algorithms for computing a Stackelberg equilibrium can be adapted to efficiently produce a Stackelberg punishment. We demonstrate an application of this idea in an experiment involving a virtual autonomous vehicle and human participants. We find that a self-driving car with a Stackelberg punishment policy discourages human drivers from bullying in a driving scenario requiring social negotiation.

In clinical care, obtaining a correct diagnosis is the first step towards successful treatment and, ultimately, recovery. Depending on the complexity of the case, the diagnostic phase can be lengthy and ridden with errors and delays. Such errors have a high likelihood to cause patients severe harm or even lead to their death and are estimated to cost the U.S. healthcare system several hundred billion dollars each year. To avoid diagnostic errors, physicians increasingly rely on diagnostic decision support systems drawing from heuristics, historic cases, textbooks, clinical guidelines and scholarly biomedical literature. The evaluation of such systems, however, is often conducted in an ad-hoc fashion, using non-transparent methodology, and proprietary data. This paper presents DC3, a collection of 31 extremely difficult diagnostic case challenges, manually compiled and solved by clinical experts. For each case, we present a number of temporally ordered physician-generated observations alongside the eventually confirmed true diagnosis. We additionally provide inferred dense relevance judgments for these cases among the PubMed collection of 27 million scholarly biomedical articles.

In e-commerce, a user tends to search for the desired product by issuing a query to the search engine and examining the retrieved results. If the search engine was successful in correctly understanding the user's query, it will return results that correspond to the products whose attributes match the terms in the query that are representative of the query's product intent. However, the search engine may fail to retrieve results that satisfy the query's product intent and thus degrading user experience due to different issues in query processing: (i) when multiple terms are present in a query it may fail to determine the relevant terms that are representative of the query's product intent, and (ii) it may suffer from vocabulary gap between the terms in the query and the product's description, i.e., terms used in the query are semantically similar but different from the terms in the product description. Hence, identifying the terms that describe the query's product intent and predicting additional terms that describe the query's product intent better than the existing query terms to the search engine is an essential task in e-commerce search. In this paper, we leverage the historical query reformulation logs of a major e-commerce retailer to develop distant-supervised approaches to solve both these problems. Our approaches exploit the fact that the significance of a term is dependent upon the context (other terms in the neighborhood) in which it is used in order to learn the importance of the term towards the query's product intent. We show that identifying and emphasizing the terms that define the query's product intent leads to a 3% improvement in ranking. Moreover, for the tasks of identifying the important terms in a query and for predicting the additional terms that represent product intent, experiments illustrate that our approaches outperform the non-contextual baselines.

[Context:] Model-based testing is an instrument for automated generation of test cases. It requires identifying requirements in documents, understanding them syntactically and semantically, and then translating them into a test model. One light-weight language for these test models are Cause-Effect-Graphs (CEG) that can be used to derive test cases. [Problem:] The creation of test models is laborious and we lack an automated solution that covers the entire process from requirement detection to test model creation. In addition, the majority of requirements is expressed in natural language (NL), which is hard to translate to test models automatically. [Principal Idea:] We build on the fact that not all NL requirements are equally unstructured. We found that 14 % of the lines in requirements documents of our industry partner contain "pseudo-code"-like descriptions of business rules. We apply Machine Learning to identify such semi-structured requirements descriptions and propose a rule-based approach for their translation into CEGs. [Contribution:] We make three contributions: (1) an algorithm for the automatic detection of semi-structured requirements descriptions in documents, (2) an algorithm for the automatic translation of the identified requirements into a CEG and (3) a study demonstrating that our proposed solution leads to 86 % time savings for test model creation without loss of quality.

Neural encoding and decoding, which aim to characterize the relationship between stimuli and brain activities, have emerged as an important area in cognitive neuroscience. Traditional encoding models, which focus on feature extraction and mapping, consider the brain as an input-output mapper without inner states. In this work, inspired by the fact that human brain acts like a state machine, we proposed a novel encoding framework that combines information from both the external world and the inner state to predict brain activity. The framework comprises two parts: forward encoding model that deals with visual stimuli and inner state model that captures influence from intrinsic connections in the brain. The forward model can be any traditional encoding model, making the framework flexible. The inner state model is a linear model to utilize information in the prediction residuals of the forward model. The proposed encoding framework can achieve much better performance on natural image identification from fMRI response than forwardonly models. The identification accuracy will decrease slightly with the dataset size increasing, but remain relatively stable with different identification methods. The results confirm that the new encoding framework is effective and robust when used for brain decoding.

Program synthesis using inputs and outputs is a fundamental problem in computer science. Towards that end, we present a framework, called NetSyn, that synthesizes programs using an evolutionary algorithm. NetSyn makes several novel contributions. First, NetSyn uses neural networks as a fitness function. This addresses the principal challenge of evolutionary algorithm: how to design the most effective fitness function. Second, NetSyn combines an evolutionary algorithm with neighborhood search to expedite the convergence process. Third, NetSyn can support a variety of neural network fitness functions uniformly. We evaluated NetSyn to generate programs in a list-based domain specific language. We compared the proposed approach against a state-of-the-art approach to show that NetSyn performs better in synthesizing programs.