Much of model-based reinforcement learning involves learning a model of an agent's world, and training an agent to leverage this model to perform a task more efficiently. While these models are demonstrably useful for agents, every naturally occurring model of the world of which we are aware---e.g., a brain---arose as the byproduct of competing evolutionary pressures for survival, not minimization of a supervised forward-predictive loss via gradient descent. That useful models can arise out of the messy and slow optimization process of evolution suggests that forward-predictive modeling can arise as a side-effect of optimization under the right circumstances. Crucially, this optimization process need not explicitly be a forward-predictive loss. In this work, we introduce a modification to traditional reinforcement learning which we call observational dropout, whereby we limit the agents ability to observe the real environment at each timestep.

Its impact is drastic and real: Youtube's AIdriven recommendation system would present sports videos for days if one happens to watch a live baseball game on the platform [1]; email writing becomes much faster with machine learning (ML) based auto-completion [2]; many businesses have adopted natural language processing based chatbots as part of their customer services [3]. AI has also greatly advanced human capabilities in complex decision-making processes ranging from determining how to allocate security resources to protect airports [4] to games such as poker [5] and Go [6]. All such tangible and stunning progress suggests that an "AI summer" is happening. As some put it, "AI is the new electricity" [7]. Meanwhile, in the past decade, an emerging theme in the AI research community is the so-called "AI for social good" (AI4SG): researchers aim at developing AI methods and tools to address problems at the societal level and improve the wellbeing of the society.

Much of model-based reinforcement learning involves learning a model of an agent's world, and training an agent to leverage this model to perform a task more efficiently. While these models are demonstrably useful for agents, every naturally occurring model of the world of which we are aware---e.g., a brain---arose as the byproduct of competing evolutionary pressures for survival, not minimization of a supervised forward-predictive loss via gradient descent. That useful models can arise out of the messy and slow optimization process of evolution suggests that forward-predictive modeling can arise as a side-effect of optimization under the right circumstances. Crucially, this optimization process need not explicitly be a forward-predictive loss. In this work, we introduce a modification to traditional reinforcement learning which we call observational dropout, whereby we limit the agents ability to observe the real environment at each timestep. In doing so, we can coerce an agent into learning a world model to fill in the observation gaps during reinforcement learning. We show that the emerged world model, while not explicitly trained to predict the future, can help the agent learn key skills required to perform well in its environment. Videos of our results available at https://learningtopredict.github.io/

Many theories, based on neuroscientific and psychological empirical evidence and on computational concepts, have been elaborated to explain the emergence of consciousness in the central nervous system. These theories propose key fundamental mechanisms to explain consciousness, but they only partially connect such mechanisms to the possible functional and adaptive role of consciousness. Recently, some cognitive and neuroscientific models try to solve this gap by linking consciousness to various aspects of goal-directed behaviour, the pivotal cognitive process that allows mammals to flexibly act in challenging environments. Here we propose the Representation Internal-Manipulation (RIM) theory of consciousness, a theory that links the main elements of consciousness theories to components and functions of goal-directed behaviour, ascribing a central role for consciousness to the goal-directed manipulation of internal representations. This manipulation relies on four specific computational operations to perform the flexible internal adaptation of all key elements of goal-directed computation, from the representations of objects to those of goals, actions, and plans. Finally, we propose the concept of `manipulation agency' relating the sense of agency to the internal manipulation of representations. This allows us to propose that the subjective experience of consciousness is associated to the human capacity to generate and control a simulated internal reality that is vividly perceived and felt through the same perceptual and emotional mechanisms used to tackle the external world.

Learning knowledge representation is an increasingly important technology applicable in many domain-specific machine learning problems. We discuss the effectiveness of traditional Link Prediction or Knowledge Graph Completion evaluation protocol when embedding knowledge representation for categorised multi-relational data in the clinical domain. Link prediction uses to split the data into training and evaluation subsets, leading to loss of information along training and harming the knowledge representation model accuracy. We propose a Clustering Evaluation Protocol as a replacement alternative to the traditionally used evaluation tasks. We used embedding models trained by a knowledge embedding approach which has been evaluated with clinical datasets. Experimental results with Pearson and Spearman correlations show strong evidence that the novel proposed evaluation protocol is pottentially able to replace link prediction.

Patterns, within the context of computational imaging, can be defined as "a more or less repeatable, discernible regularity in the spatial arrangement of a type of theme of more or less recurring objects, with possibly both radiometrical and geometrical features, sometimes referred to as elements of a collection of objects."[i] The term "radiometrical features" refers to what can be called "tone." Machines discern this portion of machine "sight" by measuring how an object responds to electromagnetic radiation. A machine blasts an object with either ultraviolet, visible, or infrared light, or all three and uses the resulting radiation to recognize patterns and compare them against known patterns.

Conflicting information in an agent's knowledge base may lead to a semantical defect, that is, a situation where it is impossible to draw any plausible conclusion. Finding out the reasons for the observed inconsistency (so-called diagnoses) and/or restoring consistency in a certain minimal way (so-called repairs) are frequently occurring issues in knowledge representation and reasoning. In this article we provide a series of first results for these problems in the context of abstract argumentation theory regarding the two most important reasoning modes, namely credulous as well as sceptical acceptance. Our analysis includes the following problems regarding minimal repairs/diagnoses: existence, verification, computation of one and enumeration of all solutions. The latter problem is tackled with a version of the so-called hitting set duality first introduced by Raymond Reiter in 1987. It turns out that grounded semantics plays an outstanding role not only in terms of complexity, but also as a useful tool to reduce the search space for diagnoses regarding other semantics.

Learning text representation is crucial for text classification and other language related tasks. There are a diverse set of text representation networks in the literature, and how to find the optimal one is a non-trivial problem. Recently, the emerging Neural Architecture Search (NAS) techniques have demonstrated good potential to solve the problem. Nevertheless, most of the existing works of NAS focus on the search algorithms and pay little attention to the search space. In this paper, we argue that the search space is also an important human prior to the success of NAS in different applications. Thus, we propose a novel search space tailored for text representation. Through automatic search, the discovered network architecture outperforms state-of-the-art models on various public datasets on text classification and natural language inference tasks. Furthermore, some of the design principles found in the automatic network agree well with human intuition.

Originally published in 1968 by Hart, Nilsson, and Raphael,2 the well-known A* search algorithm is a foundational pathfinding algorithm in computer science and artificial intelligence (AI) for traversing trees and graphs. The method provides the optimal path from the initial state to the target goal state, given the use of an admissible heuristic (must not overestimate the remaining distance to the goal). The A* algorithm is included in nearly all AI textbooks and courses worldwide. Given its widespread fame, however, there is no reliably documented evidence as to the origin of the name "A*": What does it really stand for and what does it mean? This Communications Viewpoint answers the question.

Answer Set Programming (ASP) is nowadays a dominant rule-based knowledge representation tool. Though existing ASP variants enjoy efficient implementations, generating an answer set remains intractable. The goal of this research is to define a new \asp-like rule language, 4SP, with tractable model generation. The language combines ideas of ASP and a paraconsistent rule language 4QL. Though 4SP shares the syntax of \asp and for each program all its answer sets are among 4SP models, the new language differs from ASP in its logical foundations, the intended methodology of its use and complexity of computing models. As we show in the paper, 4QL can be seen as a paraconsistent counterpart of ASP programs stratified with respect to default negation. Although model generation of well-supported models for 4QL programs is tractable, dropping stratification makes both 4QL and ASP intractable. To retain tractability while allowing non-stratified programs, in 4SP we introduce trial expressions interlacing programs with hypotheses as to the truth values of default negations. This allows us to develop a~model generation algorithm with deterministic polynomial time complexity. We also show relationships among 4SP, ASP and 4QL.