Dialogue Policy Research Articles

Making the Invisible Visible: Disability Inclusive Development in Solomon Islands

International and national level disability inclusive discourse, policy and strategy typically render invisible the institutional structures, political economies and socio-cultural power relations that constitute the lived experience of disability. The lack of robust information on disability extends these theoretical blind spots to the absence of disability in official statistics, political dialogue and social policy. Empirically grounded, inclusive research that recognises disability as embedded in globalised political economies and culturally specific power relations is required to address the theoretical, statistical and programmatic invisibility of disability. Such research will provide a solid knowledge base on which to build effective inclusive interventions.

Legacies of Lead in Charm City's Soil: Lessons from the Baltimore Ecosystem Study.

Understanding the spatial distribution of soil lead has been a focus of the Baltimore Ecosystem Study since its inception in 1997. Through multiple research projects that span spatial scales and use different methodologies, three overarching patterns have been identified: (1) soil lead concentrations often exceed state and federal regulatory limits; (2) the variability of soil lead concentrations is high; and (3) despite multiple sources and the highly heterogeneous and patchy nature of soil lead, discernable patterns do exist. Specifically, housing age, the distance to built structures, and the distance to a major roadway are strong predictors of soil lead concentrations. Understanding what drives the spatial distribution of soil lead can inform the transition of underutilized urban space into gardens and other desirable land uses while protecting human health. A framework for management is proposed that considers three factors: (1) the level of contamination; (2) the desired land use; and (3) the community’s preference in implementing the desired land use. The goal of the framework is to promote dialogue and resultant policy changes that support consistent and clear regulatory guidelines for soil lead, without which urban communities will continue to be subject to the potential for lead exposure.

Podstawowe sprzeczności w rosyjsko‑indyjskich relacjach i zmiana wektora polityki zagranicznej Federacji Rosyjskiej w ramach trójkąta Rosja–Chiny–Indie

The Pivotal Collisions of Russia‑India Relations and the Changing Vector of Russian Policy within the RIC Triangle
 This article deals with contemporary Russia‑India collisions in the field of arms trade and nuclear energy sector. The author researches principal reasons of existing problems and analyses its influence on bilateral dialogue and Russian policy within the RIC Triangle and shaping of BRICS. He concludes that Russia is going to change its traditional focus on India in favor of deepening relations with China.

Reinforcement-learning based dialogue system for human–robot interactions with socially-inspired rewards

This paper investigates some conditions under which polarized user appraisals gathered throughout the course of a vocal interaction between a machine and a human can be integrated in a reinforcement learning-based dialogue manager. More specifically, we discuss how this information can be cast into socially-inspired rewards for speeding up the policy optimisation for both efficient task completion and user adaptation in an online learning setting. For this purpose a potential-based reward shaping method is combined with a sample efficient reinforcement learning algorithm to offer a principled framework to cope with these potentially noisy interim rewards. The proposed scheme will greatly facilitate the system's development by allowing the designer to teach his system through explicit positive/negative feedbacks given as hints about task progress, in the early stage of training. At a later stage, the approach will be used as a way to ease the adaptation of the dialogue policy to specific user profiles. Experiments carried out using a state-of-the-art goal-oriented dialogue management framework, the Hidden Information State (HIS), support our claims in two configurations: firstly, with a user simulator in the tourist information domain (and thus simulated appraisals), and secondly, in the context of man–robot dialogue with real user trials.

The Dynamics of Lithuanian-Polish Strategic Partnership

The paper aims at identifying relations between the events which influence Lithuanian-Polish strategic cooperation, defining principal aspects of cooperation dynamics, and analysing recent challenges in relations between Lithuania and Poland. For the purpose of analysis the following objectives have been set: 1) to analyse the development of strategic partnership and political dialogue in bilateral relations; 2) to evaluate the importance of security, defence policy, and economic projects in cooperation between the states; 3) to assess the aspect of ethnic minorities in the context of bilateral relations. The authors of the paper seek to review the principal internal and external factors which affect bilateral cooperation between Lithuania and Poland. The following methods of analysis are used in the paper: public statements made by officials, document analysis and discourse formed by the media. The key areas of analysis are the development of political dialogue, strategic cooperation, security and defence policy, economic and energy cooperation, and questions of ethnic minorities in bilateral relations. Presently in the field there is a lack of thorough investigation of similarities and differences of strategic cooperation between Lithuania and Poland.

Nonstrict Hierarchical Reinforcement Learning for Interactive Systems and Robots

Conversational systems and robots that use reinforcement learning for policy optimization in large domains often face the problem of limited scalability. This problem has been addressed either by using function approximation techniques that estimate the approximate true value function of a policy or by using a hierarchical decomposition of a learning task into subtasks. We present a novel approach for dialogue policy optimization that combines the benefits of both hierarchical control and function approximation and that allows flexible transitions between dialogue subtasks to give human users more control over the dialogue. To this end, each reinforcement learning agent in the hierarchy is extended with a subtask transition function and a dynamic state space to allow flexible switching between subdialogues. In addition, the subtask policies are represented with linear function approximation in order to generalize the decision making to situations unseen in training. Our proposed approach is evaluated in an interactive conversational robot that learns to play quiz games. Experimental results, using simulation and real users, provide evidence that our proposed approach can lead to more flexible (natural) interactions than strict hierarchical control and that it is preferred by human users.

What We’re Talking About When We Talk About the “Civic Field” (And why we should clarify what we mean)

The field of public engagement is experiencing a harmful identity crisis. While advocates of public participation may all agree that our work relates somehow to democracy, we have not established or articulated a common vision of what that really means. This lack of clarity has dire consequences, producing rifts between academics and practitioners, community organizers and deliberative democrats, civic technologists and dialogue practitioners, policy advocates and consensus-builders. Worst of all, the lack of clarity about democracy provides no help to people who are trying to create sustainable, participatory political systems in Egypt, Thailand, Ukraine, and many other countries. None of the participatory tactics and assets we have developed will reach their full potential if we don’t admit, to ourselves and the world, their true significance: these aren’t just props for conventional processes, but building blocks for new political systems.

Gaussian Processes for POMDP-Based Dialogue Manager Optimization

A partially observable Markov decision process (POMDP) has been proposed as a dialog model that enables automatic optimization of the dialog policy and provides robustness to speech understanding errors. Various approximations allow such a model to be used for building real-world dialog systems. However, they require a large number of dialogs to train the dialog policy and hence they typically rely on the availability of a user simulator. They also require significant designer effort to hand-craft the policy representation. We investigate the use of Gaussian processes (GPs) in policy modeling to overcome these problems. We show that GP policy optimization can be implemented for a real world POMDP dialog manager, and in particular: 1) we examine different formulations of a GP policy to minimize variability in the learning process; 2) we find that the use of GP increases the learning rate by an order of magnitude thereby allowing learning by direct interaction with human users; and 3) we demonstrate that designer effort can be substantially reduced by basing the policy directly on the full belief space thereby avoiding ad hoc feature space modeling. Overall, the GP approach represents an important step forward towards fully automatic dialog policy optimization in real world systems.

A Recursive Dialogue Game for Personalized Computer-Aided Pronunciation Training

Learning languages in addition to the native language is very important for all people in the globalized world today, and computer-aided pronunciation training (CAPT) is attractive since the software can be used anywhere at any time, and repeated as many times as desired. In this paper, we introduce the immersive interaction scenario offered by spoken dialogues to CAPT by proposing a recursive dialogue game to make CAPT personalized. A number of tree-structured sub-dialogues are linked sequentially and recursively as the script for the game. The system policy at each dialogue turn is to select in real-time along the dialogue the best training sentence for each specific individual learner within the dialogue script, considering the learner's learning status and the future possible dialogue paths in the script, such that the learner can have the scores for all pronunciation units considered reaching a predefined standard in a minimum number of turns. The purpose here is that those pronunciation units poorly produced by the specific learner can be offered with more practice opportunities in the future sentences along the dialogue, which enables the learner to improve the pronunciation without having to repeat the same training sentences many times. This makes the learning process for each learner completely personalized. The dialogue policy is modeled by Markov decision process (MDP) with high-dimensional continuous state space, and trained with fitted value iteration using a huge number of simulated learners. These simulated leaners have the behavior similar to real learners, and were generated from a corpus of real learner data. The experiments demonstrated very promising results and a real cloud-based system is also successfully implemented.

Imitating Dialog Strategies Under Uncertainty

We consider human-robot interaction involving a service robot and many different users in a public environment. The task is to learn a dialog policy that deals with changing user goals, can act under uncertainty, and is easy to apply in practice. Unlike reinforcement- learning-based systems, our simulator-free approach avoids common problems such as reward tuning and state space exploration: We apply imitation learning in order to mimic an expert's behavior based on a small number of Wizard-of-Oz experiments. A dynamic Bayesian Network is used to track hidden user goals. We evaluate our approach in a simulated environment and show that by using lifelong model updates it is possible to apply the expert's policy correctly even if the user behavior changes over time.

ANÁLISE DA IMPLEMENTAÇÃO DA POLÍTICA MUNICIPAL DE EDUCAÇÃO AMBIENTAL NO MUNICÍPIO DE NATAL/RN

The overall goal of this research is to analyze the implementation of the Common Municipal Environmental Education in the city of Natal / RN (PMEAN). To achieve this goal it was necessary to identify the activities, actions, procedures, techniques, mechanisms for popular participation and financial contributions that are being used by the managers of the policy, to engage individuals and the collectivity of the scenarios of formal and non-formal about environmental education. The methodology used to analyze the strategies of implementation of the policy in question was to consult the literature on the subject, the research sites on scientific articles related to the topic, the examination of documents, practical experiences, instances of dialogue and agreement policies, programs, projects, plans, political and pedagogical; applying qualitative approach with open questionnaires on thematic context. The target of this investigation was municipal managers of Municipal and Environmental Education, elementary school teachers from public and private schools of the city of Natal, students of these schools and residents of communities the space Natal city. To obtain the results of this research was considered the validation and effectiveness of implementation strategies of PMEAN. For this reason, the study of the elements was investigated as parameters established by Law 9795/99, the National Environmental Education Politics in Brazil.

Coping with the unemployment crisis in Europe

Abstract.Since the 1970s, Europe has been plagued by a chronic unemployment crisis, which has escalated with the current Great Recession. The author discusses a number of relevant elements in this regard, including the extent to which social dialogue and social pacts have been successful in reforming labour markets and pension systems, the effect of active labour market policies and the characteristics of “resilient” economies. She concludes with the need to rethink social dialogue, social safety nets, the granting of loans – especially to SMEs – and the coordination of macroeconomic policies to increase labour demand.

Can Social Dialogue Be a Social Coordination Mechanism? Social Dialogue Policies in Chile between 1990 and 2010

This article explores social dialogue (SD) as it relates to systems theory. A brief historical overview of Chilean SD policies between 1990 and 2010 precedes analysis of employer and union leader discourse on the topic of SD. We suggest that although SD condenses positive projections, and consequently is valued by the parties involved in labor relations, concrete instances have encountered obstacles related to legitimization and institutionalization of SD. We propose that the primary factors diminishing the probability of communication are the semantics governing each party's perceptions, each party's hetero‐references, and negative expectations of its counterpart. Nevertheless, SD could become an appropriate social coordination mechanism for an increasingly complex, democratic society.Related Articles in this Politics & Policy Special Issue: Luckhurst and Zavala de Alba. 2012. “.” Politics & Policy 40 (): 735‐739. http://onlinelibrary.wiley.com/doi/10.1111/j.1747‐1346.2012.00388.x/abstract Palonen. 2012. “.” Politics & Policy 40 (): 930‐957. http://onlinelibrary.wiley.com/doi/10.1111/j.1747‐1346.2012.00389.x/abstract Zavala de Alba. 2012. .” Politics & Policy 40 (): 783‐801. http://onlinelibrary.wiley.com/doi/10.1111/j.1747‐1346.2012.00385.x/abstract Este artículo explora el Diálogo Social (DS) y su relación con la teoría de sistemas. Una breve revisión histórica de las políticas de DS chilenas entre 1990 y 2010 precede el análisis del discurso de los patrones empleadores y líderes sindicales sobre el DS. Nosotros sugerimos que aunque el DS condensa proyecciones positivas y en consecuencia es valorado por las partes involucradas en las relaciones laborales, las instancias concretas han encontrado obstáculos relacionados con la legitimación e institucionalización del DS. Proponemos que los principales factores que disminuyen la probabilidad de la comunicación son: la semántica que rodea la percepción de cada partes y las hetero‐referencias de cada partido y las expectativas negativas de cada contraparte. Sin embargo, el DS se podría convertir en un mecanismo de coordinación social apropiado para una democracia cada vez más democrática y completa.

Research for the innovation of the agri-food system in international cooperation

In the last few decades, characterized by intense environmental, landscape and socio-economic-financial changes, unexpected issues concerning the primary sector are arising and asks to researchers an urgent and deep reformulation both of the conceptual and technological tools used in the applied research. The concept of innovation is now radically changed, starting from a prevailing formula of linear top-down technology transfer, which has characterized the green revolution of the 1960s, to nowadays approach characterized by the innovation of a complex system, aimed at creating sustainable and shared opportunities through economic and institutional development. Those who now work in public or private bodies oriented to research for innovation are struggling to maintain their specific study area, but within integrated schemes where technical and scientific aspects are in interaction with organizational, institutional and political issues. Innovation can tackle several issues: new products, new technologies, new markets, new procedures (institutions) and new policies. The series of scientific and conceptual tools framed into the Agro-ecology domain seems appropriate to plan development initiatives of which the primary objective is ensuring a sustainable management of all the resources involved in agricultural production processes, while promoting food security and sovereignty, as well as protecting the rural landscape. By studying a target agro-ecosystem it is possible to identify and characterize the relationships between both the internal components and the system structures and functions at different levels of complexity of plot, farm and country, without neglecting the interactions among scientific, technological and socio-economic factors, and ultimately tending towards a science aimed at conflict resolution. Given the challenges that agricultural development is going to face in the next decades, it is indeed essential to support the planning and implementation of sound agro-ecological policies through an appropriate set of advanced tools. Particularly, a key aspect to empower all the stakeholders involved in the research, development, dissemination and application of new methods and technologies is the identification of information and communication systems that farmers and policy makers need in order to cope with fast-changing conditions in a complex system. University and research centers can act as institutional facilitators of dialogue and development policies among different stakeholders, catalyzing participation and promoting participatory problem-solving strategies for agricultural development and cooperation.

Dialogue as nuclear waste management policy: can a Swedish transparency programme legitimise a final decision on spent nuclear fuel?

The purpose of this article is to discuss Swedish nuclear waste management policy and assess whether a transparency programme has affected the policy discourse and contributed to legitimatising the forthcoming government decision on a final solution for spent nuclear fuel. The transparency programme implemented by the National Council for Nuclear Waste (NW Council) consisted of two main activities: dialogue meetings and public hearings for deliberating on different aspects of the waste issue. Lessons are drawn from an evaluation of the programme. The article concludes that the NW Council has implemented the government's legitimatisation policy by its existence and through the transparency programme, but achieving long lasting legitimacy is a complicated matter. Dialogue meetings and public hearings are reasonable measures; whether these measures are sufficient to legitimatise the final decision also depends on how the appropriate state agencies and the government justify their assessments and decisions regarding the long-term safety, security and sustainability of the Swedish Nuclear Fuel and Waste Management Company's proposed final repository. The dialogue policy and transparency programme, however, has failed to influence the key decisions which remain in the control of the powerful interests, most notable Swedish Nuclear Fuel and Waste Management Company (SKB) and the government.

Learning automata-based approach to learn dialogue policies in large state space

This paper addresses the problem of scalable optimisation of dialogue policies in speech-based conversational systems using reinforcement learning. More specifically, for large state spaces several difficulties like large tables, an account of prior knowledge and data sparsity are faced. Hence, we present an online policy learning algorithm based on hierarchical structure learning automata using eligibility trace method to find optimal dialogue strategies that cover large state spaces. The proposed algorithm is capable of deriving an optimal policy that prescribes what action should be taken in various states of conversation so as to maximise the expected total reward to attain the goal and incorporates good exploration and exploitation in its updates to improve the naturalness of human-computer interaction. The proposed model is tested using the most sophisticated evaluation framework PARADISE for accessing the travel information system.

Reinforcement learning for parameter estimation in statistical spoken dialogue systems

Reinforcement techniques have been successfully used to maximise the expected cumulative reward of statistical dialogue systems. Typically, reinforcement learning is used to estimate the parameters of a dialogue policy which selects the system's responses based on the inferred dialogue state. However, the inference of the dialogue state itself depends on a dialogue model which describes the expected behaviour of a user when interacting with the system. Ideally the parameters of this dialogue model should be also optimised to maximise the expected cumulative reward. This article presents two novel reinforcement algorithms for learning the parameters of a dialogue model. First, the Natural Belief Critic algorithm is designed to optimise the model parameters while the policy is kept fixed. This algorithm is suitable, for example, in systems using a handcrafted policy, perhaps prescribed by other design considerations. Second, the Natural Actor and Belief Critic algorithm jointly optimises both the model and the policy parameters. The algorithms are evaluated on a statistical dialogue system modelled as a Partially Observable Markov Decision Process in a tourist information domain. The evaluation is performed with a user simulator and with real users. The experiments indicate that model parameters estimated to maximise the expected reward function provide improved performance compared to the baseline handcrafted parameters.

Sample-efficient batch reinforcement learning for dialogue management optimization

Spoken Dialogue Systems (SDS) are systems which have the ability to interact with human beings using natural language as the medium of interaction. A dialogue policy plays a crucial role in determining the functioning of the dialogue management module. Handcrafting the dialogue policy is not always an option, considering the complexity of the dialogue task and the stochastic behavior of users. In recent years approaches based on Reinforcement Learning (RL) for policy optimization in dialogue management have been proved to be an efficient approach for dialogue policy optimization. Yet most of the conventional RL algorithms are data intensive and demand techniques such as user simulation. Doing so, additional modeling errors are likely to occur. This paper explores the possibility of using a set of approximate dynamic programming algorithms for policy optimization in SDS. Moreover, these algorithms are combined to a method for learning a sparse representation of the value function. Experimental results show that these algorithms when applied to dialogue management optimization are particularly sample efficient , since they learn from few hundreds of dialogue examples. These algorithms learn in an off-policy manner, meaning that they can learn optimal policies with dialogue examples generated with a quite simple strategy. Thus they can learn good dialogue policies directly from data , avoiding user modeling errors.

ISLAM – RUSSIAN ORTHODOX CHURCH RELATIONS AND THE STATE IN THE POST-COMMUNIST RUSSIA

After the fall of Soviet Union we can state a fact of a religious heritage in Russia. It is applied not only to the traditional religious institutions, like Moscow Patriarchy, but also to different heterodox religious movements. Now we can state a fact of the originally shaped religion policy of Moscow. Orthodox Christianity in Russia has one universally recognized center – Moscow Patriarchy. Its position is shared by 90% of Russian Christians. But we can’t say that the leaders have one consolidated opinion about the problems of the relations between Islam and Christianity. We can single out two groups; one can be called „For Islam” and the second „Against Islam”. Their followers have different views on the problems of proselytism, inter-religious dialogue and religion policy of the state and the foreign policy of Russia. The same, but much more multifaceted situation is in Russian Islam. Muslims in Russia don’t have any universally recognized authority. There are several organizations, regional or aspiring to the center position, authorities, sheikhs and popular homilists with their own opinions. Here we try to classify the main organizations and their views on the problem of a dialogue with Russian Orthodoxy, Christianity at all and the Russian State’s regional policy.

Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems

This paper describes a statistically motivated framework for performing real-time dialogue state updates and policy learning in a spoken dialogue system. The framework is based on the partially observable Markov decision process (POMDP), which provides a well-founded, statistical model of spoken dialogue management. However, exact belief state updates in a POMDP model are computationally intractable so approximate methods must be used. This paper presents a tractable method based on the loopy belief propagation algorithm. Various simplifications are made, which improve the efficiency significantly compared to the original algorithm as well as compared to other POMDP-based dialogue state updating approaches. A second contribution of this paper is a method for learning in spoken dialogue systems which uses a component-based policy with the episodic Natural Actor Critic algorithm. The framework proposed in this paper was tested on both simulations and in a user trial. Both indicated that using Bayesian updates of the dialogue state significantly outperforms traditional definitions of the dialogue state. Policy learning worked effectively and the learned policy outperformed all others on simulations. In user trials the learned policy was also competitive, although its optimality was less conclusive. Overall, the Bayesian update of dialogue state framework was shown to be a feasible and effective approach to building real-world POMDP-based dialogue systems.