Institut für Informatik
Refine
Year of publication
Document Type
- Article (483)
- Monograph/Edited Volume (131)
- Doctoral Thesis (88)
- Other (5)
- Preprint (4)
- Conference Proceeding (1)
- Moving Images (1)
Is part of the Bibliography
- yes (713)
Keywords
- Answer set programming (5)
- Answer Set Programming (4)
- answer set programming (3)
- Answer Set Programming (ASP) (2)
- Automata systems (2)
- Constraint Answer Set Programming (CASP) (2)
- E-learning (2)
- Prototyping (2)
- Theory (2)
- cooperating systems (2)
Institute
Recent philosophical analyses of the epistemic dimension of images in the sciences show a certain trend in acknowledging potential roles of these images beyond their merely decorative or pedagogical functions. We argue, however, that this new debate has yet paid little attention to a special type of pictures, we call ‘visual metaphor’, and its versatile heuristic potential in organizing data, supporting communication, and guiding research, modeling, and theory formation. Based on a case study of Conrad Hal Waddington’s epigenetic landscape images in biology, we develop a descriptive framework applicable to heuristic roles of various visual metaphors in the sciences.
We propose a solution based on networks of picture processors to the problem of picture pattern matching. The network solving the problem can be informally described as follows: it consists of two subnetworks, one of them extracts at each step, simultaneously, all subpictures of identical (progressively decreasing) size from the input picture and sends them to the other subnetwork which checks whether any of the received pictures is identical to the pattern. We present an efficient solution based on networks with evolutionary processors only, for patterns with at most three rows or columns. Afterward, we present a solution based on networks containing both evolutionary and hiding processors running in O(n + m + kl) computational (processing and communication) steps, for any size (n, m) of the input pic-ture and (k, l) of the pattern. From the proofs of these results, we infer that any (k, l)-local language with 1 <= k <= 3 can be decided in O(n + m + l) computational steps by networks with evolutionary processors only, while any (k, l)-local language with arbitrary k, l can be decided in O(n + m + kl) computational steps by networks containing both evolutionary and hiding processors.
Incremental Support Vector Machines (SVM) are instrumental in practical applications of online learning. This work focuses on the design and analysis of efficient incremental SVM learning, with the aim of providing a fast, numerically stable and robust implementation. A detailed analysis of convergence and of algorithmic complexity of incremental SVM learning is carried out. Based on this analysis, a new design of storage and numerical operations is proposed, which speeds up the training of an incremental SVM by a factor of 5 to 20. The performance of the new algorithm is demonstrated in two scenarios: learning with limited resources and active learning. Various applications of the algorithm, such as in drug discovery, online monitoring of industrial devices and and surveillance of network traffic, can be foreseen.
Combined optimization of spatial and temporal filters for improving brain-computer interfacing
(2006)
Brain-computer interface (BCI) systems create a novel communication channel from the brain to an output de ice by bypassing conventional motor output pathways of nerves and muscles. Therefore they could provide a new communication and control option for paralyzed patients. Modern BCI technology is essentially based on techniques for the classification of single-trial brain signals. Here we present a novel technique that allows the simultaneous optimization of a spatial and a spectral filter enhancing discriminability rates of multichannel EEG single-trials. The evaluation of 60 experiments involving 22 different subjects demonstrates the significant superiority of the proposed algorithm over to its classical counterpart: the median classification error rate was decreased by 11%. Apart from the enhanced classification, the spatial and/or the spectral filter that are determined by the algorithm can also be used for further analysis of the data, e.g., for source localization of the respective brain rhythms.
We elaborate a boundary Fourier method for studying an analogue of the Hilbert problem for analytic functions within the framework of generalised Cauchy-Riemann equations. The boundary value problem need not satisfy the Shapiro-Lopatinskij condition and so it fails to be Fredholm in Sobolev spaces. We show a solvability condition of the Hilbert problem, which looks like those for ill-posed problems, and construct an explicit formula for approximate solutions.
The performance of value classes is highly dependent on how they are represented in the virtual machine. Value class instances are immutable, have no identity, and can only refer to other value objects or primitive values and since they should be very lightweight and fast, it is important to optimize them carefully. In this paper we present a technique to detect and compress common patterns of value class usage to improve memory usage and performance. The technique identifies patterns of frequent value object references and introduces abbreviated forms for them. This allows to store multiple inter-referenced value objects in an inlined memory representation, reducing the overhead stemming from meta data and object references. Applied to a small prototype and an implementation of the Racket language, we found improvements in memory usage and execution time for several micro-benchmarks. (C) 2016 Elsevier B.V. All rights reserved.
Grammars with valuations are context-free rewriting mechanisms where the derivation process is controlled by a recursive function that evaluates strings. They have been introduced by Jurgen Dassow as models for the molecular replication process taking into account its selective character. A symbol is active in a grammar with valuation if it can be rewritten non-identically. This paper studies the effect of restricting the number of active symbols in grammars with valuations and several variants thereof to their generative power. It is investigated in which cases the number of active symbols induces infinite strict hierarchies and when the hierarchies collapse. The induced language families are compared among one another. (C) 2016 Elsevier B.V. All rights reserved.
In sub-Saharan Africa, infectious diseases and malnutrition constitute the main health problems in children, while adolescents and adults are increasingly facing cardio-metabolic conditions. Among adolescents as the largest population group in this region, we investigated the co-occurrence of infectious diseases, malnutrition and cardio-metabolic risk factors (CRFs), and evaluated demographic, socio-economic and medical risk factors for these entities. In a cross-sectional study among 188 adolescents in rural Ghana, malarial infection, common infectious diseases and Body Mass Index were assessed. We measured ferritin, C-reactive protein, retinol, fasting glucose and blood pressure. Socio-demographic data were documented. We analyzed the proportions (95% confidence interval, CI) and the cooccurrence of infectious diseases (malaria, other common diseases), malnutrition (underweight, stunting, iron deficiency, vitamin A deficiency [VAD]), and CRFs (overweight, obesity, impaired fasting glucose, hypertension). In logistic regression, odds ratios (OR) and 95% CIs were calculated for the associations with socio-demographic factors. In this Ghanaian population (age range, 14.4-15.5 years; males, 50%), the proportions were for infectious diseases 45% (95% CI: 38-52%), for malnutrition 50% (43-57%) and for CRFs 16% (11- 21%). Infectious diseases and malnutrition frequently co-existed (28%; 21-34%). Specifically, VAD increased the odds of non-malarial infectious diseases 3-fold (95% CI: 1.03, 10.19). Overlap of CRFs with infectious diseases (6%; 2-9%) or with malnutrition (7%; 3-11%) was also present. Male gender and low socio-economic status increased the odds of infectious diseases and malnutrition, respectively. Malarial infection, chronic malnutrition and VAD remain the predominant health problems among these Ghanaian adolescents. Investigating the relationships with evolving CRFs is warranted.
Irradiance from sunlight changes in a sinusoidal manner during the day, with irregular fluctuations due to clouds, and light-dark shifts at dawn and dusk are gradual. Experiments in controlled environments typically expose plants to constant irradiance during the day and abrupt light-dark transitions. To compare the effects on metabolism of sunlight versus artificial light regimes, Arabidopsis thaliana plants were grown in a naturally illuminated greenhouse around the vernal equinox, and in controlled environment chambers with a 12-h photoperiod and either constant or sinusoidal light profiles, using either white fluorescent tubes or light-emitting diodes (LEDs) tuned to a sunlight-like spectrum as the light source. Rosettes were sampled throughout a 24-h diurnal cycle for metabolite analysis. The diurnal metabolite profiles revealed that carbon and nitrogen metabolism differed significantly between sunlight and artificial light conditions. The variability of sunlight within and between days could be a factor underlying these differences. Pairwise comparisons of the artificial light sources (fluorescent versus LED) or the light profiles (constant versus sinusoidal) showed much smaller differences. The data indicate that energy-efficient LED lighting is an acceptable alternative to fluorescent lights, but results obtained from plants grown with either type of artificial lighting might not be representative of natural conditions.
Exploring one-sided communication and synchronization on a non-cache-coherent many-core architecture
(2017)
The ongoing many-core design aims at core counts where cache coherence becomes a serious challenge. Therefore, this paper discusses how one-sided communication and the required process synchronization can be realized on a non-cache-coherent many-core CPU. The Intel Single-chip Cloud Computer serves as an exemplary hardware architecture. The presented approach is based on software-managed cache coherence for MPI one-sided communication. The prototype implementation delivers a PUT performance of up to 5 times faster than the default message-based approach and reveals a reduction of the communication costs for the NAS Parallel Benchmarks 3-D fast Fourier Transform by a factor of 5. Further, the paper derives conclusions for future non-cache-coherent architectures.
The recent series 5 of the Answer Set Programming (ASP) system clingo provides generic means to enhance basic ASP with theory reasoning capabilities. We instantiate this framework with different forms of linear constraints and elaborate upon its formal properties. Given this, we discuss the respective implementations, and present techniques for using these constraints in a reactive context. More precisely, we introduce extensions to clingo with difference and linear constraints over integers and reals, respectively, and realize them in complementary ways. Finally, we empirically evaluate the resulting clingo derivatives clingo[dl] and clingo[lp] on common language fragments and contrast them to related ASP systems.
We study prediction problems in which the conditional distribution of the output given the input varies as a function of task variables which, in our applications, represent space and time. In varying-coefficient models, the coefficients of this conditional are allowed to change smoothly in space and time; the strength of the correlations between neighboring points is determined by the data. This is achieved by placing a Gaussian process (GP) prior on the coefficients. Bayesian inference in varying-coefficient models is generally intractable. We show that with an isotropic GP prior, inference in varying-coefficient models resolves to standard inference for a GP that can be solved efficiently. MAP inference in this model resolves to multitask learning using task and instance kernels. We clarify the relationship between varying-coefficient models and the hierarchical Bayesian multitask model and show that inference for hierarchical Bayesian multitask models can be carried out efficiently using graph-Laplacian kernels. We explore the model empirically for the problems of predicting rent and real-estate prices, and predicting the ground motion during seismic events. We find that varying-coefficient models with GP priors excel at predicting rents and real-estate prices. The ground-motion model predicts seismic hazards in the State of California more accurately than the previous state of the art.
Iterated finite state sequential transducers are considered as language generating devices. The hierarchy induced by the size of the state alphabet is proved to collapse to the fourth level. The corresponding language families are related to the families of languages generated by Lindenmayer systems and Chomsky grammars. Finally, some results on deterministic and extended iterated finite state transducers are established.
In this paper two new methods for the design of fault-tolerant pipelined sequential and combinational circuits, called Error Detection and Partial Error Correction (EDPEC) and Full Error Detection and Correction (FEDC), are described. The proposed methods are based on an Error Detection Logic (EDC) in the combinational circuit part combined with fault tolerant memory elements implemented using fault tolerant master-slave flip-flops. If a transient error, due to a transient fault in the combinational circuit part is detected by the EDC, the error signal controls the latching stage of the flip-flops such that the previous correct state of the register stage is retained until the transient error disappears. The system can continue to work in its previous correct state and no additional recovery procedure (with typically reduced clock frequency) is necessary. The target applications are dataflow processing blocks, for which software-based recovery methods cannot be easily applied. The presented architectures address both single events as well as timing faults of arbitrarily long duration. An example of this architecture is developed and described, based on the carry look-ahead adder. The timing conditions are carefully investigated and simulated up to the layout level. The enhancement of the baseline architecture is demonstrated with respect to the achieved fault tolerance for the single event and timing faults. It is observed that the number of uncorrected single events is reduced by the EDPEC architecture by 2.36 times compared with previous solution. The FEDC architecture further reduces the number of uncorrected events to zero and outperforms the Triple Modular Redundancy (TMR) with respect to correction of timing faults. The power overhead of both new architectures is about 26-28% lower than the TMR. (C) 2015 Elsevier Ltd. All rights reserved.
Answer Set Programming (ASP) is a powerful declarative programming paradigm that has been successfully applied to many different domains. Recently, ASP has also proved successful for hard optimization problems like course timetabling and travel allotment. In this paper, we approach another important task, namely, the shift design problem, aiming at an alignment of a minimum number of shifts in order to meet required numbers of employees (which typically vary for different time periods) in such a way that over- and understaffing is minimized. We provide an ASP encoding of the shift design problem, which, to the best of our knowledge, has not been addressed by ASP yet. Our experimental results demonstrate that ASP is capable of improving the best known solutions to some benchmark problems. Other instances remain challenging and make the shift design problem an interesting benchmark for ASP-based optimization methods.
A holistic design and verification environment to investigate driving assistance systems is presented, with an emphasis on system-on-chip architectures for video applications. Starting with an executable specification of a driving assistance application, subsequent transformations are performed across different levels of abstraction until the final implementation is achieved. The hardware/software partitioning is facilitated through the integration of OpenCV and SystemC in the same design environment, as well as OpenCV and Linux in the run-time system. We built a rapid prototyping, FPGA-based camera system, which allows designs to be explored and evaluated in realistic conditions. Using lane departure and the corresponding performance speedup, we show that our platform reduces the design time, while improving the verification efforts.
Solving problems combining task and motion planning requires searching across a symbolic search space and a geometric search space. Because of the semantic gap between symbolic and geometric representations, symbolic sequences of actions are not guaranteed to be geometrically feasible. This compels us to search in the combined search space, in which frequent backtracks between symbolic and geometric levels make the search inefficient. We address this problem by guiding symbolic search with rich information extracted from the geometric level through culprit detection mechanisms.
Learning to control a structured-prediction decoder for detection of HTTP-layer DDoS attackers
(2016)
We focus on the problem of detecting clients that attempt to exhaust server resources by flooding a service with protocol-compliant HTTP requests. Attacks are usually coordinated by an entity that controls many clients. Modeling the application as a structured-prediction problem allows the prediction model to jointly classify a multitude of clients based on their cohesion of otherwise inconspicuous features. Since the resulting output space is too vast to search exhaustively, we employ greedy search and techniques in which a parametric controller guides the search. We apply a known method that sequentially learns the controller and the structured-prediction model. We then derive an online policy-gradient method that finds the parameters of the controller and of the structured-prediction model in a joint optimization problem; we obtain a convergence guarantee for the latter method. We evaluate and compare the various methods based on a large collection of traffic data of a web-hosting service.
Answer set programming (ASP) has emerged as an approach to declarative problem solving based on the stable model semantics for logic programs. The basic idea is to represent a computational problem by a logic program, formulating constraints in terms of rules, such that its answer sets correspond to problem solutions. To this end, ASP combines an expressive language for high-level modeling with powerful low-level reasoning capacities, provided by off-the-shelf tools. Compact problem representations take advantage of genuine modeling features of ASP, including (first-order) variables, negation by default, and recursion. In this article, we demonstrate the ASP methodology on two example scenarios, illustrating basic as well as advanced modeling and solving concepts. We also discuss mechanisms to represent and implement extended kinds of preferences and optimization. An overview of further available extensions concludes the article.