SCIENCE fiction is littered with examples of intelligent computers, from HAL 9000 in "2001: A Space Odyssey" to Eddie in "The Hitchhiker's Guide to the Galaxy". One thing such fictional machines have in common is a tendency to go wrong, to the detriment of the characters in the story. Eddie obsesses about trivia, and thus puts the spacecraft he is in charge of in danger of destruction. In both cases, an attempt to build something useful and helpful has created a monster. Successful science fiction necessarily plays on real hopes and fears.

Last summer, the Defense Science Board's report on autonomy found that investing in artificial intelligence (AI) warfare is a crucial part of maintaining the United States' national security and military capability. As the report reads, "It should not be a surprise when adversaries employ autonomy against U.S. forces." In other words, AI warfare is likely on the horizon; it's just a matter of who gets there first. This immediately sparks dystopian and apocalyptic reactions from most people, who may envision a Terminator-esque system that will at some point choose to overthrow its human masters. The report concludes that "autonomy will deliver substantial operational value across an increasingly diverse array of DoD missions, but the DoD must move more rapidly to realize this value." Meaning that while the value of autonomy is clear from a military perspective, the Department of Defense has to devote more money and time to realize its full potential -- and do so quickly.

Deep models that are both effective and explainable are desirable in many settings; prior explainable models have been unimodal, offering either image-based visualization of attention weights or text-based generation of post-hoc justifications. We propose a multimodal approach to explanation, and argue that the two modalities provide complementary explanatory strengths. We collect two new datasets to define and evaluate this task, and propose a novel model which can provide joint textual rationale generation and attention visualization. Our datasets define visual and textual justifications of a classification decision for activity recognition tasks (ACT-X) and for visual question answering tasks (VQA-X). We quantitatively show that training with the textual explanations not only yields better textual justification models, but also better localizes the evidence that supports the decision. We also qualitatively show cases where visual explanation is more insightful than textual explanation, and vice versa, supporting our thesis that multimodal explanation models offer significant benefits over unimodal approaches.

Rienstra 2012) to indicate whether all and only these arguments Within the field of artificial intelligence, abstract argumentation are active, or with each spanning subtree of the argument frameworks have emerged as a useful methodology graph (Hunter 2014) to indicate that all and only the to represent and evaluate nonmonotonic logics. They allow attacks contained in that subtree are active. In all these models, to create a simple, directed graph from a defeasible knowledge an interesting question is to determine the probability base that consists of only arguments (nodes) and attacks for a set of arguments to be acceptable. A different branch (directed edges), then to identify sets of "acceptable" of research on probabilistic argumentation uses probabilities arguments in that graph, and finally to interpret these arguments' to represent the epistemic state of arguments, attacks, or sets conclusions as models in the knowledge base. In this of arguments, i.e., the belief in those elements (in terms of framework, when evaluating which arguments are acceptable acceptance). Although technically similar, this approach has in the graph, the internal structure of arguments is neglected, a completely different purpose than ours, which is the representation which accounts for the simplicity of the formalism.

Currently, much of machine learning is opaque, just like a "black box." However, in order for humans to understand, trust and effectively manage the emerging AI systems, an AI needs to be able to explain its decisions and conclusions. In this paper, I propose an argumentation-based approach to explainable AI, which has the potential to generate more comprehensive explanations than existing approaches.

End-to-end argument mining has enabled the development of new automated essay scoring (AES) systems that use argumentative features (e.g., number of claims, number of support relations) in addition to traditional legacy features (e.g., grammar, discourse structure) when scoring persuasive essays. While prior research has proposed different argumentative features as well as empirically demonstrated their utility for AES, these studies have all had important limitations. In this paper we identify a set of desiderata for evaluating the use of argument mining for AES, introduce an end-to-end argument mining system and associated argumentative feature sets, and present the results of several studies that both satisfy the desiderata and demonstrate the value-added of argument mining for scoring persuasive essays.

Text entailment, the task of determining whether a piece of text logically follows from another piece of text, has become an important component for many natural language processing tasks, such as question answering and information retrieval. For entailments requiring world knowledge, most systems still work as a "black box," providing a yes/no answer that doesn't explain the reasoning behind it. We propose an interpretable text entailment approach that, given a structured definition graph, uses a navigation algorithm based on distributional semantic models to find a path in the graph which links text and hypothesis. If such path is found, it is used to provide a human-readable justification explaining why the entailment holds. Experiments show that the proposed approach present results comparable to some well-established entailment algorithms, while also meeting Explainable AI requirements, supplying clear explanations which allow the inference model interpretation.

In this work, we apply argumentation mining techniques, in particular relation prediction, to study political speeches in monological form, where there is no direct interaction between opponents. We argue that this kind of technique can effectively support researchers in history, social and political sciences, which must deal with an increasing amount of data in digital form and need ways to automatically extract and analyse argumentation patterns. We test and discuss our approach based on the analysis of documents issued by R. Nixon and J. F. Kennedy during 1960 presidential campaign. We rely on a supervised classifier to predict argument relations (i.e., support and attack), obtaining an accuracy of 0.72 on a dataset of 1,462 argument pairs. The application of argument mining to such data allows not only to highlight the main points of agreement and disagreement between the candidates' arguments over the campaign issues such as Cuba, disarmament and health-care, but also an in-depth argumentative analysis of the respective viewpoints on these topics.

Dynamics of argumentation is the family of techniques concerned with the evolution of an argumentation framework (AF), for instance to guarantee that a given set of arguments is accepted. This work proposes Control Argumentation Frameworks (CAFs), a new approach that generalizes existing techniques, namely normal extension enforcement, by accommodating the possibility of uncertainty in dynamic scenarios. A CAF is able to deal with situations where the exact set of arguments is unknown and subject to evolution, and the existence (or direction) of some attacks is also unknown. It can be used by an agent to ensure that a set of arguments is part of one (or every) extension whatever the actual set of arguments and attacks. A QBF encoding of reasoning with CAFs provides a computational mechanism for determining whether and how this goal can be reached. We also provide some results concerning soundness and completeness of the proposed encoding as well as complexity issues.

Abstract Dialectical Frameworks (ADFs) generalize Dung's argumentation frameworks allowing various relationships among arguments to be expressed in a systematic way. We further generalize ADFs so as to accommodate arbitrary acceptance degrees for the arguments. This makes ADFs applicable in domains where both the initial status of arguments and their relationship are only insufficiently specified by Boolean functions. We define all standard ADF semantics for the weighted case, including grounded, preferred and stable semantics. We illustrate our approach using acceptance degrees from the unit interval and show how other valuation structures can be integrated. In each case it is sufficient to specify how the generalized acceptance conditions are represented by formulas, and to specify the information ordering underlying the characteristic ADF operator. We also present complexity results for problems related to weighted ADFs.