Thomas Hellström

Michele Persiani, Thomas Hellström

In Reinforcement Learning interpretability generally means to provide insight into the agent's mechanisms such that its decisions are understandable by an expert upon inspection. This definition, with the resulting methods from the literature, may however fall short for online settings where the fluency of interactions prohibits deep inspections of the decision-making algorithm. To support interpretability in online settings it is useful to borrow from the Explainable Planning literature methods that focus on the legibility of the agent, by making its intention easily discernable in an observer model. As we propose in this paper, injecting legible behavior inside an agent's policy doesn't require modify components of its learning algorithm. Rather, the agent's optimal policy can be regularized for legibility by evaluating how the policy may produce observations that would make an observer infer an incorrect policy. In our formulation, the decision boundary introduced by legibility impacts the states in which the agent's policy returns an action that has high likelihood also in other policies. In these cases, a trade-off between such action, and legible/sub-optimal action is made.

Michele Persiani, Thomas Hellstrom

When a robot is asked to verbalize its plan it can do it in many ways. For example, a seemingly natural strategy is incremental, where the robot verbalizes its planned actions in plan order. However, an important aspect of this type of strategy is that it misses considerations on what is effectively informative to communicate, because not considering what the user knows prior to explanations. In this paper we propose a verbalization strategy to communicate robot plans informatively, by measuring the information gain that verbalizations have against a second-order theory of mind of the user capturing his prior knowledge on the robot. As shown in our experiments, this strategy allows to understand the robot's goal much quicker than by using strategies such as increasing or decreasing plan order. In addition, following our formulation we hint to what is informative and why when a robot communicates its plan.

A. Sutherland, S. Bensch, T. Hellström et al.

Sentiment analysis has transitioned from classifying the sentiment of an entire sentence to providing the contextual information of what targets exist in a sentence, what sentiment the individual targets have, and what the causal words responsible for that sentiment are. However, this has led to elaborate requirements being placed on the datasets needed to train neural networks on the joint triplet task of determining an entity, its sentiment, and the causal words for that sentiment. Requiring this kind of data for training systems is problematic, as they suffer from stacking subjective annotations and domain over-fitting leading to poor model generalisation when applied in new contexts. These problems are also likely to be compounded as we attempt to jointly determine additional contextual elements in the future. To mitigate these problems, we present a hybrid neural-symbolic method utilising a Dependency Tree-LSTM's compositional sentiment parse structure and complementary symbolic rules to correctly extract target-sentiment-cause triplets from sentences without the need for triplet training data. We show that this method has the potential to perform in line with state-of-the-art approaches while also simplifying the data required and providing a degree of interpretability through the Tree-LSTM.

Maitreyee Tewari, Thomas Hellström, Suna Bensch

This paper describes a recently initiated research project aiming at supporting development of computerised dialogue systems that handle breaches of conversational norms such as the Gricean maxims, which describe how dialogue participants ideally form their utterances in order to be informative, relevant, brief, etc. Our approach is to model dialogue and norms with co-operating distributed grammar systems (CDGSs), and to develop methods to detect breaches and to handle them in dialogue systems for verbal human-robot interaction.

Thomas Hellström, Virginia Dignum, Suna Bensch

In public media as well as in scientific publications, the term \emph{bias} is used in conjunction with machine learning in many different contexts, and with many different meanings. This paper proposes a taxonomy of these different meanings, terminology, and definitions by surveying the, primarily scientific, literature on machine learning. In some cases, we suggest extensions and modifications to promote a clear terminology and completeness. The survey is followed by an analysis and discussion on how different types of biases are connected and depend on each other. We conclude that there is a complex relation between bias occurring in the machine learning pipeline that leads to a model, and the eventual bias of the model (which is typically related to social discrimination). The former bias may or may not influence the latter, in a sometimes bad, and sometime good way.