Test (all options have to be dealt with) Flashcards
Richtig (Green) – Falsch (Red) – Who knows? (Blue/Black)
- Machine learning → incremental learner?
Right (Green):
a) algorithm that learn iteratively with each provided data input
b) decreases error rate
- Least square method → linear?
Right (Green):
a) linear classifier
b) linear regression
Black:
predictive modelling
- Hysteresis refers to what?
Right (Green):
a) Dynamic system – depends on history – take one of several different states with the same parameter values.
False (Right);
b) Bandwidth of possible system
c) Rapid phase transition
- Netlogo behaviour space function?
Right (Green):
b) perform parameter sweeps for statistic evaluation
c) exclude particular behaviours of system
- Power-law distribution?
Right: a) interaction among distributed instances
False: b) also seen in body size
Right: c) found in wealth distribution
- Bush fire example?
Right:
a) rapid phase
c) non-linear change
- Cusp catastrophe?
Right:
- History (hysteresis)
- One of the 7 elementary catastrophes described by the catastrophe theory
- Shannon information → carry information?
Right:
- more info than others are more often occurring
- rare signs have higher surprise value
- Preferential attachment?
Right:
- Power law distribution
- This model generates these networks by a process of “preferential attachment”, in which new network members prefer to make a connection to the more popular existing members.
- Machine learning → data discrimination?
Right: S. 55: Linear discrimination (or segmentation) of data refers to a set of methods in machine learning which are able to distinguish - and to some extent predict membership in - different classes by way of a linear combination - a weighted sum - of the attributes of the data. This discrimination is usually given in the form of a straight separation line - the linear discriminant. • Linear regression • Linear classification • Predictive modelling • Support Vector Machine • Logistic regression
- Jacobian matrix → classify equilibria, defines links between hysteresis, nodes, allows, determine
Right:
- Used to classify equilibria in dynamic systems
Wrong:
- Defines links between nodes in a graph?
- Hysteresis?
- Jacobian matrix, define the determinant?
Right:
Det=a11a22 – a12a21
Trace= a11+a22
- Python-Code → What does it? What is matplotlib?
Right:
Shows a squared list of integers
- Intinerant or strange attractor?
An attractor is an invariant subset of a phase space towards which dynamical systems tend to evolve in discrete time regardless of their starting conditions. Simple attractors can be fixed points, sets of points, limit cycles or manifolds. More interesting attractors are “strange”, “chaotic” or “itinerant” attractors, which span an array of possible states in which a dynamical system can roam around without repeating itself.
The following interactive model presents some examples of “strange attractors”.
The Lorenz attractor - butterfly effect
The Peter de Jong attractor
The Hénon attractor
The Rössler attractor
The Standard attractor
- A positive Lyapunov exponent is what?
Right:
- Indicates a dynamic equilibrium within a bandwidth of the phase space
- Positive = diverge
- Negative = converge
- N-body problem illustrates?
Right:
Possibility of deterministic chaos in dynamical systems
- Term frequency – Inverse Document Frequency (TF-IDF) matrix?
Right:
Rare words are weighted more heavily than often used words.
- Unstable fixed point equilibrium is indicated if the real parts of the eigenvalues of the Jacobian matrix are……?
Find answer to that question!
- Machine learning-accuracy indicates?
Right:
- Accuracy indicating the degree of closeness of a classification to the actual (true) values
- Accuracy = TP + TN / (TP + FP + TN + FN)
- Using k-nearest neighbour’s classifier it is better to consider → small, odd or large groups?
Right:
Odd
