Logo des Repositoriums

Künstliche Intelligenz 29(4) - November 2015

Autor*innen mit den meisten Dokumenten  

Auflistung nach:

Neueste Veröffentlichungen

1 - 10 von 16
  • Zeitschriftenartikel
    Special Issue on Autonomous Learning
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Hammer, Barbara; Toussaint, Marc
  • Zeitschriftenartikel
    Beyond Manual Tuning of Hyperparameters
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Hutter, Frank; Lücke, Jörg; Schmidt-Thieme, Lars
    The success of hand-crafted machine learning systems in many applications raises the question of making machine learning algorithms more autonomous, i.e., to reduce the requirement of expert input to a minimum. We discuss two strategies towards this goal: (1) automated optimization of hyperparameters (including mechanisms for feature selection, preprocessing, model selection, etc) and (2) the development of algorithms with reduced sets of hyperparameters. Since many research directions (e.g., deep learning), show a tendency towards increasingly complex algorithms with more and more hyperparamters, the demand for both of these strategies continuously increases. We review recent hyperparameter optimization methods and discuss data-driven approaches to avoid the introduction of hyperparameters using unsupervised learning. We end in discussing how these complementary strategies can work hand-in-hand, representing a very promising approach towards autonomous machine learning.
  • Zeitschriftenartikel
    Learning Feedback in Intelligent Tutoring Systems
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Gross, Sebastian; Mokbel, Bassam; Hammer, Barbara; Pinkwart, Niels
    Intelligent Tutoring Systems (ITSs) are adaptive learning systems that aim to support learners by providing one-on-one individualized instruction. Typically, instructing learners in ITSs is build on formalized domain knowledge and, thus, the applicability is restricted to well-defined domains where knowledge about the domain being taught can be explicitly modeled. For ill-defined domains, human tutors still by far outperform the performance of ITSs, or the latter are not applicable at all. As part of the DFG priority programme “Autonomous Learning”, the FIT project has been conducted over a period of 3 years pursuing the goal to develop novel ITS methods, that are also applicable for ill-defined problems, based on implicit domain knowledge extracted from educational data sets. Here, machine learning techniques have been used to autonomously infer structures from given learning data (e.g., student solutions) and, based on these structures, to develop strategies for instructing learners.
  • Zeitschriftenartikel
    Open-EASE: A Cloud-Based Knowledge Service for Autonomous Learning
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Tenorth, Moritz; Winkler, Jan; Beßler, Daniel; Beetz, Michael
    We present Open-EASE, a cloud-based knowledge base of robot experience data that can serve as episodic memory, providing a robot with comprehensive information for autonomously learning manipulation tasks. Open-EASE combines both robot and human activity data in a common, semantically annotated knowledge base, including robot poses, object information, environment models, the robot’s intentions and beliefs, as well as information about the actions that have been performed. A powerful query language and inference tools support reasoning about the data and retrieving information based on semantic queries. In this paper, we focus on applications of Open-EASE in the context of autonomous learning.
  • Zeitschriftenartikel
    Statistical Relational Artificial Intelligence: From Distributions through Actions to Optimization
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Kersting, Kristian; Natarajan, Sriraam
    Statistical Relational AI—the science and engineering of making intelligent machines acting in noisy worlds composed of objects and relations among the objects—is currently motivating a lot of new AI research and has tremendous theoretical and practical implications. Theoretically, combining logic and probability in a unified representation and building general-purpose reasoning tools for it has been the dream of AI, dating back to the late 1980s. Practically, successful statistical relational AI tools enable new applications in several large, complex real-world domains including those involving big data, natural text, social networks, the web, medicine and robotics, among others. Such domains are often characterized by rich relational structure and large amounts of uncertainty. Logic helps to faithfully model the former while probability helps to effectively manage the latter. Our intention here is to give a brief (and necessarily incomplete) overview and invitation to the emerging field of Statistical Relational AI from the perspective of acting optimally and learning to act.
  • Zeitschriftenartikel
    Training Restricted Boltzmann Machines
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Fischer, Asja
    Restricted Boltzmann Machines (RBMs), two-layered probabilistic graphical models that can also be interpreted as feed forward neural networks, enjoy much popularity for pattern analysis and generation. Training RBMs however is challenging. It is based on likelihood maximization, but the likelihood and its gradient are computationally intractable. Therefore, training algorithms such as Contrastive Divergence (CD) and learning based on Parallel Tempering (PT) rely on Markov chain Monte Carlo methods to approximate the gradient. The presented thesis contributes to understanding RBM training methods by presenting an empirical and theoretical analysis of the bias of the CD approximation and a bound on the mixing rate of PT. Furthermore, the thesis improves RBM training by proposing a new transition operator leading to faster mixing Markov chains, by investigating a different parameterization of the RBM model class referred to as centered RBMs, and by exploring estimation techniques from statistical physics to approximate the likelihood. Finally, an analysis of the representational power of deep belief networks with real-valued visible variables is given.
  • Zeitschriftenartikel
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015)
  • Zeitschriftenartikel
    Interview with Werner von Seelen
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Paaßen, Benjamin
  • Zeitschriftenartikel
    Online Learning of Bipedal Walking Stabilization
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Missura, Marcell; Behnke, Sven
    Bipedal walking is a complex whole-body motion with inherently unstable dynamics that makes the design of a robust controller particularly challenging. While a walk controller could potentially be learned with the hardware in the loop, the destructive nature of exploratory motions and the impracticality of a high number of required repetitions render most of the existing machine learning methods unsuitable for an online learning setting with real hardware. In a project in the DFG Priority Programme Autonomous Learning, we are investigating ways of bootstrapping the learning process with basic walking skills and enabling a humanoid robot to autonomously learn how to control its balance during walking.
  • Zeitschriftenartikel
    Autonomous Learning of State Representations for Control: An Emerging Field Aims to Autonomously Learn State Representations for Reinforcement Learning Agents from Their Real-World Sensor Observations
    (KI - Künstliche Intelligenz: Vol. 29, No. 4, 2015) Böhmer, Wendelin; Springenberg, Jost Tobias; Boedecker, Joschka; Riedmiller, Martin; Obermayer, Klaus
    This article reviews an emerging field that aims for autonomous reinforcement learning (RL) directly on sensor-observations. Straightforward end-to-end RL has recently shown remarkable success, but relies on large amounts of samples. As this is not feasible in robotics, we review two approaches to learn intermediate state representations from previous experiences: deep auto-encoders and slow-feature analysis. We analyze theoretical properties of the representations and point to potential improvements.