The TPC algorithm. by Rahul Biswas (14117086)

Swarm intelligence algorithms for multi-objective IMP: Step-by-step improvement by Francisco Muñoz (9455441)

“…</div><div><br></div><div>positional arguments: </div><div> file </div><div><br></div><div>optional arguments: </div><div> -h, --help: show this help message and exit </div><div> -a: read file input contents as an Adjacency Matrix </div><div> -r: reverse nodes order, reads (b,a,w) instead of (a,b,w) </div><div> -d D: line separator to use on file parsing </div><div> -t T: number of times to execute this solver </div><div> -i I: number of metaheuristic iterations per execution </div><div> -q Q: fixed quota, use 0 = floor(n/2)+1 </div><div> --shape SHAPE: shape functions for binarization - list of implemented shape functions: s2, s2_neg, v2, v4 </div><div> --sym: consider graph as symmetric instead of directed </div><div> --folder FOLDER: output folder </div><div> --seed SEED: use custom seed for metaheuristic pseudo RNG </div><div> --spread SPREAD: use specific influence spread model funcion - list of implemented models: LT, IC </div><div> --sum SUM: adds an extra value to all edge's weight --mh MH: metaheuristic to use - list of implemented metaheuristics: 1 (Swarm3) </div><div> --tagsfile TAGSFILE: use first row as node tags instead of using plurality criteria </div><div> --notags: do not use first row as node tags, instead tags will be calculated </div><div> --prune: nodes with (1) outdegree = 0 and indegree > 0, and (2) with outdegree = 1 and neighbor's outdegree > 0 will be excluded </div><div> --prunelowdeg: nodes with low outdegree or degree will be excluded </div><div> --excludenodes EXCLUDENODES: nodes to skip, must be comma separated </div><div> --includeindegzero: forces to include nodes with indegree = 0 on all executions </div><div> --nodepthcriteria: particles will not tie off with spread depth in case of having same fitness and spread </div>…”

Example performance of the smooth-index algorithm. by Alexander Borst (32619)

Subjects:

Python scripts used to model CPU errors by Poisson distribution density mass function and to draw graph of Poisson events (needs Matplot, SciPy, PysimpleGUI and NumPy) by Santhosh Kumar Rajamani (13023678)

The real NGD values, one-step ahead training set predictions, and multi-step ahead test set predictions using the AI-AFTER algorithm for Kayseri. by Burak Alakent (463512)

Matlab code of a function used in S2 File. by Hiroko Kato Solvang (5718250)

Great machine learning performance with multiple classification algorithms. by Sydney C. Weiser (15383531)

Python implementation of the Trajectory Adaptive Multilevel Sampling algorithm for rare events and improvements by Pascal Wang (10130612)

“…In `main.py`, the parameters for the TAMS algorithm are specified (trajectory time, time step, score function, number of particles, type of score threshold, maximum number of iterations, noise level etc.). …”

Network structure of perception fusion algorithm. by Weilin Wu (13736)

Subjects:

Reconstruction errors and their variability due to stochastic steps in the algorithms. by Hanane Omichessan (4991552)

Statistical results of various algorithms. by ZeSheng Lin (20501356)

Subjects:

Images of partial benchmark functions. by ZeSheng Lin (20501356)

Subjects:

Flowchart of improved FOX optimization algorithm. by Mahmood A. Jumaah (22272371)

Subjects:

Analysis of the limit cases of the sampling function. by Greta Grassmann (12418506)

Subjects:

Predictive Mixing for Density Functional Theory (and Other Fixed-Point Problems) by L. D. Marks (1949689)

Subjects:

Performance of the smooth-index algorithm for matrices with different densities. by Alexander Borst (32619)

Subjects:

Algorithm of the brightness scale calibration experiment. by Krzysztof Petelczyc (3954203)

“…<p>In the algorithm, the following variables were used: “I” denotes the current luminous intensity of the reference diode, “inc” denotes the current difference between reference and target diode luminous intensity; “cnt” is the current number of performed trials, while “correct” is a counter of correct answers in cnt trials, both of them are counted separately for every settings of I and inc. …”

Wavelet transform and scaling function. by Menaa Nawaz (14304920)

The case of LLSKSO with the value of the weight coefficient <i>w</i> in steps of 0.1. by Ruyi Dong (9038174)

Results of the application of different clustering algorithms to average functional connectivity from healthy subjects. by Francisco Páscoa dos Santos (16510676)

“…<p>A) Resulting cluster inertia from applying the k-means algorithm described in the methods to empirical averaged functional connectivity from healthy subjects, with different numbers of clusters. …”