New paradigm of FPGA-based computational intelligence from surveying the implementation of DNN accelerators

In the 1990s, a new paradigm of science characterized by uncertainty, nonlinearity, and irreversibility and tackling complex problems was generally recognized by the academic community. In this new paradigm, traditional analytical methods are ineffectual, and there is recognition of the need to explore new methods to solve the more flexible, more robust system problems. In 1994 the first Computational Intelligence Conference in Orlando, Florida, US, first combined three different areas, smart neural networks, fuzzy systems and genetic algorithms, not only because the three have many similarities, but also because a properly combined system of the three is more effective than a system generated by one single technical field. Various theories and approaches of computational intelligence including neural computing, fuzzy computing and evolutional computing are comprehensively introduced in this chapter.

Download Full-text

Solar Filaments as Tracers of Subsurface Processes

International Astronomical Union Colloquium ◽

10.1017/s0252921100064459 ◽

2000 ◽

Vol 179 ◽

pp. 177-183

Author(s):

D. M. Rust

Keyword(s):

Magnetic Fields ◽

Magnetic Flux ◽

Coronal Mass Ejections ◽

Interplanetary Space ◽

Intermediate Stage ◽

Coronal Plasma ◽

Magnetic Helicity ◽

The Sun ◽

New Paradigm ◽

Solar Filaments

AbstractSolar filaments are discussed in terms of two contrasting paradigms. The standard paradigm is that filaments are formed by condensation of coronal plasma into magnetic fields that are twisted or dimpled as a consequence of motions of the fields’ sources in the photosphere. According to a new paradigm, filaments form in rising, twisted flux ropes and are a necessary intermediate stage in the transfer to interplanetary space of dynamo-generated magnetic flux. It is argued that the accumulation of magnetic helicity in filaments and their coronal surroundings leads to filament eruptions and coronal mass ejections. These ejections relieve the Sun of the flux generated by the dynamo and make way for the flux of the next cycle.

Download Full-text

Does the Motor System Facilitate Spatial Imagery?

Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie ◽

10.1026/0049-8637/a000175 ◽

2017 ◽

Vol 49 (3) ◽

pp. 127-137 ◽

Cited By ~ 1

Author(s):

Markus Krüger ◽

Horst Krist

Keyword(s):

Motor System ◽

Third Graders ◽

External Support ◽

New Paradigm ◽

Perceptual Condition ◽

Imagery Condition ◽

Passive Rotation ◽

Perception Condition ◽

Motor Condition ◽

Dynamic Imagery

Abstract. Recent studies have ascertained a link between the motor system and imagery in children. A motor effect on imagery is demonstrated by the influence of stimuli-related movement constraints (i. e., constraints defined by the musculoskeletal system) on mental rotation, or by interference effects due to participants’ own body movements or body postures. This link is usually seen as qualitatively different or stronger in children as opposed to adults. In the present research, we put this interpretation to further scrutiny using a new paradigm: In a motor condition we asked our participants (kindergartners and third-graders) to manually rotate a circular board with a covered picture on it. This condition was compared with a perceptual condition where the board was rotated by an experimenter. Additionally, in a pure imagery condition, children were instructed to merely imagine the rotation of the board. The children’s task was to mark the presumed end position of a salient detail of the respective picture. The children’s performance was clearly the worst in the pure imagery condition. However, contrary to what embodiment theories would suggest, there was no difference in participants’ performance between the active rotation (i. e., motor) and the passive rotation (i. e., perception) condition. Control experiments revealed that this was also the case when, in the perception condition, gaze shifting was controlled for and when the board was rotated mechanically rather than by the experimenter. Our findings indicate that young children depend heavily on external support when imagining physical events. Furthermore, they indicate that motor-assisted imagery is not generally superior to perceptually driven dynamic imagery.

Download Full-text

Self-Prioritization Beyond Perception

Experimental Psychology (formerly Zeitschrift für Experimentelle Psychologie) ◽

10.1027/1618-3169/a000307 ◽

2015 ◽

Vol 62 (6) ◽

pp. 415-425 ◽

Cited By ~ 21

Author(s):

Sarah Schäfer ◽

Dirk Wentura ◽

Christian Frings

Keyword(s):

The Self ◽

Stimulus Category ◽

Abstract Concept ◽

Conceptual Level ◽

New Paradigm ◽

Stimulus Features ◽

Relevant Word ◽

Word Shape

Abstract. Recently, Sui, He, and Humphreys (2012) introduced a new paradigm to measure perceptual self-prioritization processes. It seems that arbitrarily tagging shapes to self-relevant words (I, my, me, and so on) leads to speeded verification times when matching self-relevant word shape pairings (e.g., me – triangle) as compared to non-self-relevant word shape pairings (e.g., stranger – circle). In order to analyze the level at which self-prioritization takes place we analyzed whether the self-prioritization effect is due to a tagging of the self-relevant label and the particular associated shape or due to a tagging of the self with an abstract concept. In two experiments participants showed standard self-prioritization effects with varying stimulus features or different exemplars of a particular stimulus-category suggesting that self-prioritization also works at a conceptual level.

Download Full-text