Connecting the dots between discoveries in neuroscience(neuroplasticity), psychoneuroimmunology(the brain-immune loop) and user experience (gadget rub-off) indicate the nature of our time spent with gadgets is a vector in human health - mentally, socially and physically. The positive design of our interactions with devices therefore can have a positive impact on economy, civilization and society. Likewise, the absence of design that encourages positive interaction may encourage undesirable behaviors. Much like the architecture of physical spaces and buildings, the consequences of the architecture of the 21stcentury conversation between man and machine may last generations. AI and the Internet of Things are primary vectors for positive and negative impacts of technology.  We describe a growing body of co-discoveries occurring across a variety of disciplines that support the argument for human sciences in technology design.

Research in neurally-based machine (i.e. computational) systems is expanding. “Reverse-engineered” models of brain-like structures are viable candidates for developing increasing complexification (via generatively encoded “intelligence”) that could instantiate some form of consciousness – albeit not identical to human consciousness. This essay posits how such trajectories could lead to the iterative development of “machine sentience” and addresses issues of what “machine consciousness” might mean for: 1) the ways that humans regard such machine entities as “beings” and/or “persons”, and 2) philosophical, ethical and socio-legal positions which might need to be adapted to guide and govern human treatment of, and interactions with such entities. Herein, I argue that neuroethics contributes crucial insights and viable tools to any meaningful approach to this topic (in synergy with extant discourse in “robo-ethics”). As the fields of neuro- and cognitive science, and computational engineering become increasingly convergent, so too must the philosophical and ethical approaches that can – and should – be employed to direct what convergent science may create. The speed and breadth of such technological development are such that neuroethical address and engagement of these issues and questions must be equivalently paced and iterative, so as to retain preparatory value.

In this paper, we present an approach to mine large-scale knowledge graphs to discover inference paths for query expansion in NLIDB (Natural Language Interface to Databases). Addressing this problem is important in order for NLIDB applications to effectively handle relevant concepts in the domain of interest that do not correspond to any structured fields in the target database. We also present preliminary observations on the performance of our approach applied to Freebase, and conclude with discussions on next steps to further evaluate and extend our approach.

For people constrained to picture based communication, the expression of interest in a question answering (QA) or information retrieval (IR)scenario is highly limited. Traditionally, alternative and augmentative communication (AAC) methods (such as gestures and communication boards) are utilised. But only few systems allow users to produce whole utterances or sentences that consist of multiple words; work to generate them automatically is a promising direction in the big data context.In this paper, we provide a dedicated access method for the open-domain QA and IR context. We propose a method for the user to search for additional symbols to be added to the communication board in real-time while using access to big data sources and context based filtering when the desired symbol is missing. The user can select a symbol that is associated with the desired concept and the system searches for images on the Internet - here, in Wikipedia - with the purpose of retrieving an appropriate symbol or picture. Querying for candidates is performed by estimating semantic relatedness between text fragments using explicit semantic analysis (ESA).

We describe the KELVIN system for extracting entities and relations from large text collections and its use in the TAC Knowledge Base Population Cold Start task run by the U.S. National Institute of Standards and Technology. The Cold Start task starts with an empty knowledge base defined by an ontology or entity types, properties and relations. Evaluations in 2012 and 2013 were done using a collection of text from local Web and news to de-emphasize the linking entities to a background knowledge bases such as Wikipedia. Interesting features of KELVIN include a cross-document entity coreference module based on entity mentions, removal of suspect intra-document conference chains, a slot value consolidator for entities, the application of inference rules to expand the number of asserted facts and a set of analysis and browsing tools supporting development.

We propose a learning agent that first learns concepts in an integrated, cross-modal manner, and then uses these as the semantics model to map language. We consider an abstract model for the action of throwing, modeling the entire trajectory. From a large set of throws, we take the trajectory images and and the throwing parameters. These are mapped jointly onto a low-dimensional non-linear manifold. Such models improve with practice, and can be used as the starting point for real-life tasks such as aiming darts or recognizing throws by others. How can such models can be used in learning language? We consider a set of videos involving throwing and rolling actions. These actions are analyzed into a set of contrastive semantic classes based on agent, action, and the thrown object (trajector). We obtain crowdsourced commentaries for these videos (raw text) from a number of adults. The learner attempts to associate labels using contrastive probabilities for the semantic class. Only a handful of high-confidence words are found, but the agent starts off with this partial knowledge. These are used to learn incrementally larger syntactic patterns, initially for the trajector, and eventually for full agent-trajector-action sentences. We demonstrate how this may work for two completely different languages - English and Hindi, and also show how rudiments of agreement, synonymy and polysemy are detected.

In this summary report I give an account of research conducted over the last two years, showing the suitability and the advantages of applying computational analogy-engines in the analysis and design of analogy-based methods and tools in teaching and education. This overview constitutes the conclusion of the first phase of a multi-stage effort trying to introduce computational models of analogy also to education and the learning sciences, thus opening up these fields to computational tools and methods not only on an instrumental level, but also in analytical, conceptual, and design-oriented studies. I locate the "analogy-engines in the classroom" research program within the bigger schemes of studying human creativity and computational creativity, provide an introduction to the theoretical underpinnings of the endeavor, and revisit three worked out case studies serving as proofs of the feasibility of the overall approach.

The problem of transferring skills to hyper-redundant system requires the design of new motion primitive representations that can cope with multiple sources of noise and redundancy, and that can dynamically handle perturbations in the environment. One way is to take inspiration from invertebrate systems in nature to seek for new versatile representations of motion/behavior primitives for continuum robots. In particular, the incredibly varied skills achieved by the octopus can guide us toward the design of such robust encoding scheme. This abstract presents our ongoing work that aims at combining statistical machine learning, dynamical systems and stochastic optimization to study the problem of transferring skills to a flexible surgical robot (STIFF-FLOP) composed of 2 modules with constant curvatures. The approach is tested in simulation by imitation and self-refinement of an octopus reaching motion.

The goal of transfer is to use knowledge obtained by solving one task to improvea robot's (or software agent's) performance in future tasks. In general, we do not expect this to work; for transfer to be feasible, there must be something in common between the source task(s) and goal task(s). The question at the core of the transfer learning enterprise is therefore: what makes two tasks related?, or more generally, how do you define a family of related tasks? Given a precise definition of how a particular family of tasks is related, we can formulate clear optimizationmethods for selecting source tasks and determining what knowledge should be imported from the source task(s), and how it should be used in the target task(s). This paper describes one model that has appeared in several different research scenarios where an agent is faced with afamily of tasks that have similar, but not identical, dynamics (or reward functions). For example, a human learning to play baseball may, over the course of their career,be exposed to several different bats, each with slightly different weights and lengths.A human who has learned to play baseball well with one bat would be expected to be able to pick up any similar bat and use it.Similarly, when learning to drive a car, one may learn in more than one car, and then be expected to be able to drive any make and model of car (within reasonablevariations) with little or no relearning. These examples are instances of exactly the kind of flexible, reliable,and sample-efficient behavior that we should be aiming to achieve in robotics applications. One way to model such a family of tasks is to posit that they are generated by asmall set of latent parameters (e.g., the length and weight of the bat, or parametersdescribing the various physical properties of the car's steering system and clutch) thatare fixed for each problem instance (e.g., for each bat, or car), but are not directlyobservable by the agent. Defining a distributionover these latent parameters results in a family of related tasks, and transferis feasible to the extent that the number of latent variables is small, the task dynamics(or reward function) vary smoothly with them, and to the extent to which they can eitherbe ignored or identified using transition data from the task.This model has appeared under several different names in the literature; we refer to it as a hidden-parameterMarkov decision process (or HIP-MDP).

We address two domains of skill transfer problems encountered by an autonomous robot: within-domain adaptation and cross-domain transfer. Our aim is to provide skill representations which enable transfer in each problem classification. As such, we explore two approaches to skill representation which address each problem classification separately. The first representation, based on mimicking, encodes the full demonstration and is well suited for within-domain adaptation. The second representation is based on imitation and serves to encode a set of key points along the trajectory, which represent the goal points most relevant to the successful completion of the skill. This representation enables both within-domain and cross-domain transfer. A planner is then applied to these constraints, generating a domain-specific trajectory which addresses the transfer task.