Aula x: Tı́tulo da Aula

Nome da Disciplina, xxxxxxx

Marcelo O. Ribeiro1
1marcelo.or@ufma.br, Universidade Federal do Maranhão,UFMA
15 de julho de 2024

T
his latex template provides some instruc-
tions on how to format your assignment
report. Feel free to use your favorite la-
tex editor. Follow the structure below if it suits
your needs.
1 Introduction
A short introduction to the problem.
You may want to list your contributions using
the command itemize.
add the entries to the file literature.bib. Below are
some examples of how to refer to related works.
The reader should get a copy of Bishop’s book on
pattern recognition and machine learning (Bishop,
2006) from our library. Also highly recommended
are the books of Murphy (2012) e Williams e Ras-
mussen (2006). Also make sure to refer to all
artwork like Figure 1.
Another example of actively citing authors is,
e.g., Bishop (2006) discusses many elementary
machine learning concepts.

• First item in a list

• Second item in a list 2.2 Pseudocode and algorithms
• Third item in a list
Do not list complete python routines in the
methods section. Instead, briefly present your
2 Methods method or algorithm and list only the essential
part of your code. Below is an example.
Something about the used methods.
1 # ∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗
Define your equations using the align environ- 2 # The Sigmoid A c t i v a t i o n F u n c t i o n
ment. or using eqnarray. 3 class ActivationFunction :
4 def _ _ i n i t _ _ ( s e l f ) :

A11 A21
 5 p r i n t ( " Constructor of the
A= (1) base a c t i v a t i o n f u n c t i o n
A21 A22
")
6

7 def transform ( s e l f ,
y = w1 + w2 x1 + w3 x2 + · · · + wD+1 xD ,
input_vector ) :
y = xT w. (2) 8 r e t u r n 1/(1+np . exp(−1∗
input_vector ) )
2.1 Referencing related work
Large code excerpts should be listed in the Ap-
You may want to use Google Scholar to find related pendix in a single column format to enhance the
scientific publications. Download the bib files and readability.

Page 1 of 4
 Aula x: Título da Aula

2.3 You may add more structure to your
report
Your assignment report should comply with the fol-
lowing structure. Feel free to add latex structure
elements like
\paragraph*{...}, \subsubsection*{...}
• The font size of text in figures should be equal
to the figure caption font size. E.g., in MA-
TLAB use a font size of 18 − 24 for figures.
Use the command
\begin{figure*}[t] ... \end{figure*}

\begin{itemize}..., \begin{enumerate}, etc. to print double column figures.

3 Results
Something about your results. Some plots would
be also nice.

3.1 Formatting instructions for equati-

ons
Use the notation defined in our on-
line slides defined in pages 8-11 in
https://docs.google.com/presentation/d/
1mtpjSyuoBwhZxc8QUI1PUfMhyyXAXsXNXNqHsLSk-AQ/
edit?usp=sharing. In particular, Figura 1: Illustration of an figure generated using
python code in the appendix. Make sure that
• Use single letters for scalars, i.e., a ∈ R1 . all axis labels and legend texts are readable.
• Vectors are written in bold, i.e., w ∈ R10 . The font size should be close to the the font
• Matrices are written in bold capital letters, size of this caption.
i.e., M ∈ R{10 × 10} .
• Use subscripts to denote elements in vectors,
i.e., wi is part of w = [w1 , w2 , . . . , wi ].
• Denote functions by single lower case letters, 3.3 Tables are important
i.e., x = f (y) which implements the transfor-
mation f (y) : R1 → R1 . You may want to use some online latex table
• You may use subscripts for functions, i.e., x = generators like https://www.tablesgenerator.
ffwd (y) which implements the forward kine- com/. Below are some examples.
matics transformation. Or y = finv (x) which
Tabela 1: Example table
implements the inverse kinematics transfor-
mation.
Name
• Denote continuous distribution using the
lower case p(x) term and use the symbol N (.) First Name Last Name Grade
to denote Gaussian distributions. John Doe 7.5
Richard Miles 5
3.2 Formatting instructions for figures
When creating figures,
• Make sure that all axis have labels. 3.4 Compiling the latex file
• Always add a legend (images may be excepti-
ons). You need to execute the pdflatex command to
• Use different line colors, e.g., in MATLAB use generate a .pdf file. To compile the bibliography,
the colormap Lines. you need to run bibtex. You may execute these
• Use a minimum line width of 2 and different commands in a terminal or using your favorite
line styles. latex editor.

Page 2 of 4
 Aula x: Título da Aula

4 Conclusion
Something about the problems of the method you
used and give hints on how to solve them.

Bibliography
Bishop, Christopher M (2006). Pattern recognition
and machine learning. springer.
Murphy, Kevin P (2012). Machine learning: a pro-
babilistic perspective. MIT press.
Williams, Christopher KI e Carl Edward Rasmus-
sen (2006). Gaussian processes for machine lear-
ning. Vol. 2. 3. MIT press Cambridge, MA.

Page 3 of 4
 APPENDIX
Add all relevant information to reimplement your algorithms or to reproduce the derivations to the
appendix.

Adding code
You can also directly print your source code using the command shown below.
1 # ∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗
2 # T h i s s c r i p t d e m o n s t r a t e s how a s i m p l e f i g u r e can be
3 # c r e a t e d u s i n g t h e package m a t p l o t l i b .
4 #
5 # The example was developed by Honghu Xue , a PhD s t u d e n t
6 # o f Univ.− P r o f . Dr . Elmar Rueckert , 2022
7 # ∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗∗
8 import numpy as np
9 from m a t p l o t l i b import p y p l o t as p l t
10

11 f i g = p lt . figure ()
12 x = np . l i n s p a c e ( 0 , 500 , 500)
13 y_1 = np . l o g ( x )
14 s t d _ y _ 1 = 0.02 ∗x
15 p l t . p l o t ( x , y_1 , c o l o r = ’ b ’ , alpha = 0 . 6 , l a b e l= ’ method 1 ’ )
16 p l t . f i l l _ b e t w e e n ( x , y_1−std_y_1 , y_1+s t d _ y _ 1 , c o l o r = ’ b ’ , alpha = 0 . 1 )
##539 c a f
17

18 y_2 = x ∗ 0 . 1
19 s t d _ y _ 2 = 0.01 ∗x
20 p l t . p l o t ( x , y_2 , c o l o r = ’ r ’ , alpha = 0 . 6 , l a b e l= ’ method 2 ’ )
21 p l t . f i l l _ b e t w e e n ( x , y_2−std_y_2 , y_2+s t d _ y _ 2 , c o l o r = ’ r ’ , alpha = 0 . 1 )
##539 c a f
22

23

24 #p l . t i t l e ( ’ E p i s o d i c s p i l l a g e a v e r a g e ’ )
25 p l t . xlabel ( ’ training episodes ’ )
26 p l t . y l a b e l ( ’ e p s o d i c reward ’ )
27 p l t . legend ( l o c= ’ upper l e f t ’ )
28 p l t . show ( )
29

30 f i g . s a v e f i g ( " example p l o t . svg " , format=" svg " )

Additional Results
Often additional tables and figures are presented in the appendix.

Template by Chair of Cyber-Physical-Systems, Montanuniversität Leoben, Austria Page 4 of 4