Automated in - vitro cell detection and counting ha ve been a key theme for artificial and intelligent biological analysis such as biopsy, drug analysis and decease diagnosis. Along with the rapid development of microfluidics and lab - on - chip technolog ies, in - vitro live cell analysis has be en one of the critical task s for both research and industry communities. However, it is a great challenge to obtain and then predict the precis e information of liv e cells from numerous microscopic videos and images. In this paper, we investigated in - vitro detection of white blood cell s using deep neural networks, and discuss ed how state - of - the - art machine learning techniques could fulfil the needs of medical diagnos is. The approach we used in this study wa s based on Faster Region - based Convolutional Neural Networks (Faster RCNNs), and a transfer learning process wa s applied to apply this technique to the microscopic detection of blood cell s . Our experimental results demonstrated that fast and efficient analysis of blood cell s via automated microscopic imaging can achieve much better accuracy and faster speed than the conventionally applied methods, implying a promising future of this technology to be applied to the microfluidic point - of - care medical devices .

It consists of 30k turns plus 10k annotated SQL queries, obtained from a Wizard-of-Oz (WOZ) collection of 3k dialogues querying 200 complex DBs spanning 138 domains. Each dialogue simulates a real-world DB query scenario with a crowd worker as a user exploring the DB and a SQL expert retrieving answers with SQL, clarifying ambiguous questions, or otherwise informing of unanswerable questions. When user questions are answerable by SQL, the expert describes the SQL and execution results to the user, hence maintaining a natural interaction flow. CoSQL introduces new challenges compared to existing task-oriented dialogue datasets: (1) the dialogue states are grounded in SQL, a domain-independent executable representation, instead of domain-specific slot-value pairs, and (2) because testing is done on unseen databases, success requires generalizing to new domains. CoSQL includes three tasks: SQL-grounded dialogue state tracking, response generation from query results, and user dialogue act prediction. We evaluate a set of strong baselines for each task and show that CoSQL presents significant challenges for future research. The dataset, baselines, and leaderboard will be released at https:// yale-lily.github.io/cosql .

This paper presents an exploratory study to understand the relationship between a humans' cognitive load, trust, and anthropomorphism during human-robot interaction. To understand the relationship, we created a \say{Matching the Pair} game that participants could play collaboratively with one of two robot types, Husky or Pepper. The goal was to understand if humans would trust the robot as a teammate while being in the game-playing situation that demanded a high level of cognitive load. Using a humanoid vs. a technical robot, we also investigated the impact of physical anthropomorphism and we furthermore tested the impact of robot error rate on subsequent judgments and behavior. Our results showed that there was an inversely proportional relationship between trust and cognitive load, suggesting that as the amount of cognitive load increased in the participants, their ratings of trust decreased. We also found a triple interaction impact between robot-type, error-rate and participant's ratings of trust. We found that participants perceived Pepper to be more trustworthy in comparison with the Husky robot after playing the game with both robots under high error-rate condition. On the contrary, Husky was perceived as more trustworthy than Pepper when it was depicted as featuring a low error-rate. Our results are interesting and call further investigation of the impact of physical anthropomorphism in combination with variable error-rates of the robot.

Motion Planning Explorer: Visualizing Local Minima using a Local-Minima Tree Andreas Orthey, Benjamin Fr esz, Marc Toussaint Abstract -- We present an algorithm to visualize local minima in a motion planning problem, which we call the motion planning explorer . The input to the explorer is a planning problem, a sequence of lower-dimensional projections of the configuration space, a cost functional and an optimization method. The output is a local-minima tree, which is interactively grown based on user input. We show the motion planning explorer to faithfully capture the structure of four real-world scenarios. I NTRODUCTION In motion planning, we develop algorithms to move robots from an initial configuration to a desired goal configuration. Such algorithms are essential for manufacturing, autonomous flight, computer animation or protein folding [14]. Most motion planning algorithms are black-box algorithms 1 . A user inputs a goal configuration and the algorithm returns a motion. In real-world scenarios, however, black-box algorithms are problematic. There is no way to guide or prevent motions. Humans users cannot visualize the internal mechanism of the algorithm.

Many question answering (QA) tasks only provide weak supervision for how the answer should be computed. For example, TriviaQA answers are entities that can be mentioned multiple times in supporting documents, while DROP answers can be computed by deriving many different equations from numbers in the reference text. In this paper, we show it is possible to convert such tasks into discrete latent variable learning problems with a precomputed, task-specific set of possible "solutions" (e.g. different mentions or equations) that contains one correct option. We then develop a hard EM learning scheme that computes gradients relative to the most likely solution at each update. Despite its simplicity, we show that this approach significantly outperforms previous methods on six QA tasks, including absolute gains of 2--10%, and achieves the state-of-the-art on five of them. Using hard updates instead of maximizing marginal likelihood is key to these results as it encourages the model to find the one correct answer, which we show through detailed qualitative analysis.

Prior access to domain knowledge could significantly improve the performance of a reinforcement learning agent. In particular, it could help agents avoid potentially catastrophic exploratory actions, which would otherwise have to be experienced during learning. In this work, we identify consistently undesirable actions in a set of previously learned tasks, and use pseudo-rewards associated with them to learn a prior policy. In addition to enabling safe exploratory behaviors in subsequent tasks in the domain, these priors are transferable to similar environments, and can be learned off-policy and in parallel with the learning of other tasks in the domain. We compare our approach to established, state-of-the-art algorithms in a grid-world navigation environment, and demonstrate that it exhibits a superior performance with respect to avoiding unsafe actions while learning to perform arbitrary tasks in the domain. We also present some theoretical analysis to support these results, and discuss the implications and some alternative formulations of this approach, which could also be useful to accelerate learning in certain scenarios.

As reinforcement learning (RL) achieves more success in solving complex tasks, more care is needed to ensure that RL research is reproducible and that algorithms herein can be compared easily and fairly with minimal bias. RL results are, however, notoriously hard to reproduce due to the algorithms' intrinsic variance, the environments' stochasticity, and numerous (potentially unreported) hyper-parameters. In this work we investigate the many issues leading to irreproducible research and how to manage those. We further show how to utilise a rigorous and standardised evaluation approach for easing the process of documentation, evaluation and fair comparison of different algorithms, where we emphasise the importance of choosing the right measurement metrics and conducting proper statistics on the results, for unbiased reporting of the results.

Knowledge graphs are important resources for many artificial intelligence tasks but often suffer from incompleteness. In this work, we propose to use pre-trained language models for knowledge graph completion. We treat triples in knowledge graphs as textual sequences and propose a novel framework named Knowledge Graph Bidirectional Encoder Representations from Transformer (KG-BERT) to model these triples. Our method takes entity and relation descriptions of a triple as input and computes scoring function of the triple with the KG-BERT language model. Experimental results on multiple benchmark knowledge graphs show that our method can achieve state-of-the-art performance in triple classification, link prediction and relation prediction tasks.

This paper formally compares some central notions from two well-known formalisms for rule-based argumentation, DeLP and ASPIC+. The comparisons especially focus on intuitive adequacy and inter-translatability, consistency, and closure properties. As for differences in the definitions of arguments and attack, it turns out that DeLP's definitions are intuitively appealing but that they may not fully comply with Caminada and Amgoud's rationality postulates of strict closure and indirect consistency. For some special cases, the DeLP definitions are shown to fare better than ASPIC+. Next, it is argued that there are reasons to consider a variant of DeLP with grounded semantics, since in some examples its current notion of warrant arguably has counterintuitive consequences and may lead to sets of warranted arguments that are not admissible. Finally, under some minimality and consistency assumptions on ASPIC+ arguments, a one-to-many correspondence between ASPIC+ arguments and DeLP arguments is identified in such a way that if the DeLP warranting procedure is changed to grounded semantics, then DeLP notion of warrant and ASPIC+'s notion of justification are equivalent. This result is proven for three alternative definitions of attack.

AI has achieved remarkable mastery over games such as Chess, Go, and Poker, and even Jeopardy, but the rich variety of standardized exams has remained a landmark challenge. Even in 2016, the best AI system achieved merely 59.3% on an 8th Grade science exam challenge. This paper reports unprecedented success on the Grade 8 New York Regents Science Exam, where for the first time a system scores more than 90% on the exam's non-diagram, multiple choice (NDMC) questions. In addition, our Aristo system, building upon the success of recent language models, exceeded 83% on the corresponding Grade 12 Science Exam NDMC questions. The results, on unseen test questions, are robust across different test years and different variations of this kind of test. They demonstrate that modern NLP methods can result in mastery on this task. While not a full solution to general question-answering (the questions are multiple choice, and the domain is restricted to 8th Grade science), it represents a significant milestone for the field.