Natural Language Processing NLP for Requirements Engineering A Systematic Mapping Study

This article surveys the landscape of NLP4RE research to understand the state of the art and identify open problems . 140 NLP techniques, 66 NLP tools and 25 NLP resources are extracted from the selected studies . 130 new tools have been proposed to support a range of linguistic analysis tasks, but there is little evidence of adoption in the long term, although some industrial applications have been published .…

Robots in the Danger Zone Exploring Public Perception through Engagement

Public perceptions of Robotics and Artificial Intelligence (RAI) are important in the acceptance, uptake, government regulation and research funding of this technology . Recent research has shown that the public’s understanding of RAI can be negative or inaccurate . In this paper, wedescribe our first iteration of a high throughput in-person public engagement activity .…

Bias in Machine Learning What is it Good for

In public media as well as in scientific publications, the term ‘bias’ is used in conjunction with machine learning in many different contexts . This paper proposes a taxonomy of these differentmeanings, terminology, and definitions . The former bias may or may not influence the latter, in a sometimes bad, and sometime good way .…

A Survey on Conversational Recommender Systems

Recommender systems are software applications that help users to find itemsof interest in situations of information overload . The interest in CRShas significantly increased in the past few years . This development is mainly due to the significant progress in the area of natural language processing, the emergence of new voice-controlled home assistants, and the increased use ofchatbot technology .…

Work in Progress Temporally Extended Auxiliary Tasks

Predictive auxiliary tasks have been shown to improve performance in numerousreinforcement learning works, however, this effect is still not wellunderstood . The primary purpose of the work presented here is to investigate the impact that an auxiliary task’s prediction timescale has on the agent’s policy performance .…

Learning Sparse Rewarded Tasks from Sub Optimal Demonstrations

Self-Adaptive Imitation Learning (SAIL) can achieve near optimal performance given only a limited number of sub-optimaldemonstrations for highly challenging sparse reward tasks . SAIL bridges theadvantages of IL and RL to reduce the sample complexity substantially, by exploiting sup-optimal demonstrations and efficiently exploring theenvironment to surpass the demonstrated performance .…

Counterfactual Multi Agent Reinforcement Learning with Graph Convolution Communication

We consider a fully cooperative multi-agent system where agents cooperate to maximize a system’s utility in a partial-observable environment . We propose anarchitecture that allows for communication among agents and tailors the system’s reward for each individual agent . We evaluate our method on a range of tasks, demonstrating the advantage of marrying communication with credit assignment .…

A New Challenge Approaching Tetris Link with AI

This paper focuses on a new game, TetrisLink, a board game that is still lacking any scientific analysis . Tetris Link has a large branching factor, hampering a traditional heuristic planning approach . We document our approach and report on their performance in a tournament.…

You can do RLAs for IRV

The City and County of San Francisco, CA, has used Instant Runoff Voting for some elections since 2004 . This report describes the first ever process pilot of Risk Limiting Audits for IRV, for the San Francisco DistrictAttorney’s race in November, 2019 .…

Ontology based Interpretable Machine Learning for Textual Data

In this paper, we introduce a novel interpreting framework that learns aninterpretable model based on an ontology-based sampling technique to explainagnostic prediction models . Different from existing approaches, our algorithm considers contextual correlation among words, described in domain knowledgeontologies, to generate semantic explanations .…

Proceedings 9th International Conference on Quantum Simulation and Quantum Walks

This volume contains a selection of papers presented at the 9th in a series of international conferences on Quantum Simulation and Quantum Walks . During this event, we worked on the development of theories based upon quantumwalks and quantum simulation models, in order to solve interrelated problems such as the simulation of standard quantum field theory, quantum gravity and quantum gravity .…

Two shot Spatially varying BRDF and Shape Estimation

Traditional optimization-based approaches often need a largenumber of images taken from multiple views in a controlled environment . We propose anovel deep learning architecture with a stage-wise estimation of shape and SVBRDF . The previous predictions guide each estimation, and a joint refinementnetwork later refines both SV BRDF and shape .…

SoftSMPL Data driven Modeling of Nonlinear Soft tissue Dynamics for Parametric Humans

SoftSMPL is a learning-based method to model realistic soft-tissuedynamics as a function of body shape and motion . Datasets to learn such taskare scarce and expensive to generate, which makes training models prone to tooverfitting . At the core of our method there are three key contributions that enable us to model highly realistic dynamics and better generalizationcapabilities than state-of-the-art methods, while training on the same data .…

Proceedings of the 12th International Workshop on Programming Language Approaches to Concurrency and Communication cEntric Software

The International Workshop on ProgrammingLanguage Approaches to Concurrency- and Communication-cEntric Software (PLACES) is dedicated to work in this area . The workshop offers a forum for researchers from different fields to exchange new ideas about these challenges to modern programming, where concurrency and distribution are the norm rather than a marginal concern .…

Crafty Efficient HTM Compatible Persistent Transactions

Crafty employs a novel technique called nondestructive undo logging that leverages commodity HTM to control persist ordering . Crafty outperforms state-of-the-art prior work under low contention, and performs competitively under high contention, says the authors . The paper introduces Crafty, a new approach for ensuring consistency andatomicity on persistent memory operations using commodity hardware using hardware transactional memory (HTM) capabilities, while incurring low overhead costs .…

Streaming Temporal Graphs Subgraph Matching

We present a high-level language for describing temporal subgraphs of interest, the Streaming Analytics Language (SAL) SAM programs are translated into C++ code that is run in parallel on a cluster . SAM is able to scale to 128 nodes or 2560 cores, while Apache Flink has max throughput with 32nodes and degrades thereafter .…

An Exploratory Study of Writing and Revising Explicit Programming Strategies

Knowledge sharing plays a crucial role throughout all software application development activities . When programmers learn and share through media like Stack overflow, GitHub, Meetups, videos, discussion forums, wikis, and blogs, every developer benefits . However, there is one kind of knowledge that developers share far less often: strategic knowledge for how to approach programming problems .…

Interactive Evolution and Exploration Within Latent Level Design Space of Generative Adversarial Networks

Generative Adversarial Networks (GANs) are an emerging form of indirectencoding . Latent Variable Evolution (LVE) has recently been applied to game levels . However, it is hard for objective scores to capture level features that are appealing to players . The tool also allows for direct exploration of the latent dimensions, and allows users to play discovered levels .…

Automated Configuration of Negotiation Strategies

Bidding and acceptance strategies have a substantial impact on the outcome ofnegotiations in scenarios with linear additive and nonlinear utility functions . By empoweringautomated negotiating agents using automated algorithm configuration, we obtain a flexible negotiation agent that can be configured automatically for a richspace of opponents and negotiation scenarios .…

Personal Health Knowledge Graphs for Patients

Existing patient data analytics platforms fail to incorporate information that has context, is personal, and topical to patients . For a recommendationsystem to give a suitable response to a query, it should consider personal information about the patient’s history, including but not limited to their preferences, locations, and choices that are currently applicable to them .…

A macro agent and its actions

Integrated information theory (IIT) offers a quantitative account of causation based on a set of causalprinciples, including notions such as causal specificity, composition, andirreducibility . Here, we demonstrate this frameworkby example of a simulated agent, equipped with a small neural network, that forms a maximum of $\Phi$ at a macro scale .…

Explosive Proofs of Mathematical Truths

Mathematical proofs are both paradigms of certainty and some of the mostexplicitly-justified arguments that we have in the cultural record . Their very explicitness, however, leads to a paradox, because their probability of error grows exponentially as the argument expands .…

Learning to Ask Medical Questions using Reinforcement Learning

Given a masked vector of input features, areinforcement learning agent iteratively selects certain features to beunmasked . The algorithm makes use of a novel environment setting,responding to a non-stationary Markov Decision Process . Applying our method to a national survey dataset, we show that it outperforms strongbaselines when requiring the prediction to be made based on a small number of features, but is also highly more interpretable .…

Mimicking Evolution with Reinforcement Learning

Evolution gave rise to human and animal intelligence here on Earth . We argue that the path to developing artificial human-like-intelligence will passthrough mimicking the evolutionary process in a nature-like simulation . InNature, there are two processes driving the development of the brain: evolutionand learning .…

Augmented Q Imitation Learning AQIL

Traditional deep reinforcement learning takes a significant time before the machine starts to converge to an optimal policy . This paper proposesAugmented Q-Imitation-Learning, a method by which deep . reinforcement learning can be accelerated by applying Q-imitation-learning as the initial .training…

Will we ever have Conscious Machines

The distinction of whether something is really self-aware or merely aclever program that pretends to do so cannot be answered without access to the mechanism’s inner workings . Many important algorithmic steps towards machines with a core consciousness have already been devised .…