I presented our paper Conformity and Trust in Multi-Party vs. Individual Human-Robot Interaction at the Trust session on Wednesday, September 18th.

Doctoral Consortium

I presented my Ph.D. studies on Innovative Language Learning Using AI and Robot Environment at the Doctoral Consortium on Sunday, September 15th.

Interactive Poster

Doctoral Consortium — IVA 2024 by Alireza M. Kamelabad

The second dog is a yoked control — literally “yoked,” harnessed to the first dog’s behavior the way a pair of oxen shares a crossbar. It’s one of the more elegant tricks in experimental design, and, it turns out, one of the more contested. Working through why it works, and where it breaks, is a useful lesson in experimental design generally.

What “yoked” actually buys you

A garden-variety control group just goes without the treatment. That’s not good enough here, because the thing being tested — whether an animal has control over an outcome — can’t be separated from the outcome itself by simply withholding it. If you gave the control dog no shocks at all, and it later behaved differently from the shocked dog, you’d have shown that shock does something, which nobody doubted. You wouldn’t know whether it was the shock or the helplessness driving the later passivity.

Yoking solves this by giving the control subject the exact same physical stimulus, on the exact same schedule, as its paired partner — the only thing that differs is whether the subject’s own behavior had any say in it. The full version of this logic is usually run as a three-way (triadic) comparison: an escapable-shock group, a yoked inescapable-shock group, and a no-shock group as a baseline.

flowchart TB
    classDef escapable fill:#166E5A,stroke:#166E5A,color:#fff
    classDef yoked fill:#D14E24,stroke:#D14E24,color:#fff
    classDef baseline fill:#5b5b5b,stroke:#5b5b5b,color:#fff

    A["Escapable group<br/>presses panel → shock ends"] -->|"shock onset & offset<br/>copied in real time"| B["Yoked group<br/>identical shock, no working panel"]
    C["No-shock group<br/>never shocked at all"]

    class A escapable
    class B yoked
    class C baseline

Fig. 1 — the triadic design: two groups get physically identical shocks and differ only in who controls them; the third gets none, as a baseline.

Because the escapable and yoked groups receive physically identical treatment, any difference between them afterward can’t be attributed to the shock itself — only to whether the shock was contingent on the animal’s own behavior. That’s the whole trick.

The original case: learned helplessness

That dog experiment is Martin Seligman and Steven Maier’s 1967 study, and it’s the paper that introduced the term “learned helplessness.”¹ Dogs that could press a panel to end shock later learned to escape shock in a new apparatus without trouble. Dogs yoked to them — same shocks, no working panel — later just lay down and took it, even once escape became trivially available. Seligman and Maier’s original account was that the yoked dogs had learned that nothing they did mattered, and that this learning suppressed the motivation to try.

That account held for about fifty years, and then Maier and Seligman revisited it themselves in a 2016 retrospective — and reversed the central mechanism.² Passivity, they now argue, is the default, unlearned response to a prolonged aversive event, driven by serotonin activity in the dorsal raphe nucleus. What the escapable-shock dogs had wasn’t a passivity they avoided learning — it was an active, learned inhibition of that default passivity, generated by the ventromedial prefrontal cortex the moment it detects that outcomes are controllable. In their words, “there is nothing in the brain that is selectively turned on by a lack of control, only something that turns things off when there is the presence of control.” The yoked design was never in question — it’s still exactly how you’d isolate this effect — but what it was shown to be isolating changed. The paper also notes that inescapable shock reproduces eight of the nine DSM diagnostic symptoms of depression, which is part of why the paradigm became a standard animal model in depression research.

The logic travels: ulcers, without any learning at all

The same yoked structure produces a strikingly similar result outside the realm of learning and memory altogether. In a 1968 study, Joseph Weiss ran rats through a comparable pairing: one rat could avoid or shorten shock with a coping response, its yoked partner received identical shock on an identical schedule with no working response, and a third rat got no shock. The yoked “helpless” rats developed substantially more gastric ulceration than their paired counterparts — despite receiving, again, the exact same physical stressor.³ The result argues that it’s the psychological experience of control, not the raw quantity of aversive stimulation, driving the physiological damage. Weiss also turned up a wrinkle worth keeping in mind: within the coping group, rats that made more responses developed more ulceration, not less — a reminder that “having control” and “exercising it a lot” aren’t the same variable, and a yoked design only cleanly isolates the one it was built to isolate.

Not just shocks, and not just animals

Nothing about the yoked logic requires an aversive stimulus, and the design shows up well outside the shock literature. Developmental psychologists studying infant learning use a gentler version of exactly the same structure: a ribbon ties a 2–3-month-old infant’s ankle to a mobile overhead, so kicking makes the mobile move. The infant’s kick rate roughly triples within minutes as they learn the contingency. In the yoked control condition, a different infant’s mobile moves on the same schedule as their paired partner’s — same visual stimulation, same timing, no contingency — and yoked infants show none of the learning.⁴ It’s the same experimental question — does behavior producing an effect matter, independent of the effect itself — asked with a toy instead of a shock panel.

The design hasn’t dated, either. A 2020 human study built a “translational” version of the original triadic design to study uncontrollable stress in people directly: an escapable-stress group, a yoked group receiving identical stress with the escape response disabled, and a no-stress control, used to study how perceived controllability shapes stress and depression risk in humans.⁵

The catch hiding inside the design

Here’s where the story gets less tidy. In 1964, Robert Church published a statistical objection to the yoked control design that experimental psychologists have been arguing about ever since.⁶ The yoked subject’s data isn’t statistically independent of its partner’s — by construction, the yoked animal’s entire stimulus schedule is a copy of whatever the master animal happened to do. If the master animals differ from each other for reasons that have nothing to do with the manipulation — one is just more anxious, or more active, or reacts more strongly to shock — that individual variation gets transmitted wholesale into the yoked animal’s environment. Church showed that this correlated-error structure can produce apparent group differences at a rate well above chance even when there’s no real effect of controllability at all — the design can manufacture “significant” results out of nothing but between-subject noise in the master group.

A follow-up paper by Kimmel and Terrant in 1968 took the critique seriously enough to propose a fix: a reciprocal yoked design, where each pair of subjects serves as both master and yoked partner across two different conditioned stimuli, so any individual bias one animal contributes gets balanced out rather than one-directional.⁷ It’s a partial patch, not a full solution — and it’s telling that, decades on, methods papers are still finding new bias sources in yoked comparisons.

So does it still hold up?

Mostly, yes — with a caveat about how much weight any single yoked comparison should carry. Church’s objection doesn’t mean learned helplessness, or Weiss’s ulcer result, are statistical artifacts; both have been replicated many times, across labs, strains, and eventually species, including the 2016 mechanistic account that traced the effect down to specific brain circuitry. But it does mean a lone yoked-pair comparison is weaker evidence on its own than it looks. What holds up is the pattern that keeps showing up when a yoked design is cross-checked against something else — a different paradigm, a different species, a molecular mechanism, a human neuroimaging study — rather than any single experiment in isolation.

That’s really the design’s legacy: not a flawless tool, but a genuinely clever one, honest enough about its own weak spot that the field has spent sixty years arguing about it in public. Next time you read that some group of animals, or infants, or people “had no control” in an experiment, it’s worth asking who they were yoked to — and what that pairing might be quietly carrying along with it.

Photo: an operant-conditioning (“Skinner box”) chamber set up for a pigeon-conditioning demonstration, Universidad Nacional Mayor de San Marcos, 1998 — the general class of apparatus behind the classic lever/panel-press experiments described above. Photo by Dtarazona, public domain, via Wikimedia Commons.

Your are most probably writing your thesis in LaTeX, on overleaf. I comment your thesis in three different ways:

Specific Comments

I use Overleaf’s review feature to add comments about a specific instance when there is something at the spot that is not clear to me. You need to either answer the comment, or modify the text to reflect the comment in a way that resolves the matter I have pointed out in the comment. You can reply the comment if it is not clear to you. If you are sure about your change and think that it is resolved, then you can choose “Resolve” option under the comment to close it.

General Comments and Suggestions

There might be systematic suggestions that would have several instances in your writing. For example specific grammatical mistake that you have repeated. Or a comment about your referencing style. For such matters, I comment on the first few instances that I see and after that I will use the todonotes package of LaTeX to add a comment in the margin or inline. I probably also create a list of todos somewhere at the beginning of the document. When you are done resolving them, you can put a comment on it using overleaf’s commenting system, or if you have questions about it, you can ask.

You do not need to delete these \todo commands from your document. If they are resolved simply use \usepackage[disable]{todonotes} at the preamble of your LaTeX document to disable them in your final version.

Useful LaTeX Package: Check out todonotes package. I find it very useful in many cases.

Corrections

Sometimes I might find a mistake in your thesis that needs to be corrected and it does not require your attention. I will correct it directly in the document. For such matters I use “track changes” option of Overleaf. You can see the changes I have made and accept or reject them. If you have questions about the change, you can ask me.

Common Mistakes

There are a few common mistakes that I have encountered master students make. I think it is a good idea to list them here so that you can avoid them before starting to write your thesis.

Use of Informal Language

Your thesis is a formal academic document. You should avoid using informal or casual language in your thesis. This includes contractions, slang, and colloquial expressions. A common informal structure is the use of short forms like “don’t”, “can’t”, “won’t”, etc. You should use the full form of these words in your thesis. Examples:

Use Acurate, Precise, and Unambiguous Language

Be careful with the use of adjectives and adverbs when writing an academic piece. Lots of adjectives and adverbs imply subjective interpretation and their measure differ from person to person. They are incalculable, unquantifiable, and immeasurable. Therefore, they can distort the actual message you want to convey. For example:

The results show a significantly better learning outcome of the participants in group A compared to group B (what is significantly better? How much better?).

The results show a $10%$ increase in the word retention of the participants in group A compared to group B (this is a quantifiable measure).

A more conceptual example of how adjectives can differ in meaning is:

• I am “very happy” to announce that we have a new colleague in our team.
• I am “very happy” that I won the lottery of $5 million.

In the two sentences above, both are very happy, but one is much happier than the other. Also, in the second one, me thinking about it might imagine a person crying from happiness for winning the money, while you might imagine a person jumping around. Now can you say whether the happiness I am imagining that person has is the same amount of happiness you are imagining for her? Maybe my very happy is happier than your very happy. That is why, using such terms in scientific writing is dangerous:

The fail-safe system of the seat eject mechanism is “extremely reliable”. For this aircraft it “almost never” fails.

Now I am sitting at the corner thinking ok, I guess “extremely reliable” means like 99.9999% reliable which makes me happy to use their system. Also, I would say “almost never” when it is a one in a million chance. Now imagine the author of this sentence is a person who is very optimistic and always sees the glass half full. He might say “extremely reliable” when it is 95% reliable and “almost never” when it is 1 in 100. Now I am not happy to use their system. I hope you get the gist of it.

First Person Pronoun

In almost all the cases, the master thesis is done by a single person (you). You do the practical part and also the writing. You are the sole author and beneficiary of the thesis. Therefore, when you need to use a pronoun to refer to the author, you should use the first person singular pronoun “I”. It is true that I might help you with some stages in your thesis, but I am your supervisor and it is my job to do so. Therefore I should not be credited in your thesis as an author. :)

Balance Passive and Active Voice

Conventionally, there is a preference for the passive voice in academic writing. However, the passive voice can make your writing unclear and difficult to read. Therefore, my personal preference is to use the active voice whenever possible. The active voice is more direct and concise. It is also more engaging and easier to read. Examples:

However, there are cases where the passive voice is more appropriate. For example, when the focus is on the action rather than the actor. Examples:

• It is widely believed that…
• The results show that…
• It has been shown that…
• The data suggest that…

Use of Abbreviations

When you introduce an abbreviation, you should first write the full form of the abbreviation followed by the abbreviation in parentheses. After that, you can use the abbreviation in the rest of the document. For example:

Useful LaTeX Packages: acronym and glossaries

Back up your claims! Either by reference or by proof (data)

When you make a claim in your thesis, you should back it up with evidence. This evidence can be a reference to a scientific paper, a book, or a reliable website. It can also be data that you have collected or analyzed. If you make a claim or a statement without evidence, it is considered an opinion. Opinions are not valid in academic writing. You should always back up your claims with evidence. Example:

Claim by Opinion : Electric cars are better for the environment than gasoline cars. (it sounds obvious, but without experimentation or data, it is just an opinion. Which means I can claim otherwise and you cannot do anything about it. :D)

Evidence : Electric cars produce CO2 emissions only during their production phase. After that, they are emission-free. Gasoline cars produce CO2 emissions throughout their lifetime (Smith, 2020). (In order to oppose this statement, now I need to at least provide counter-evidence or find a problem in the analysis or the quality of Smith’s study.)

Consistent Referencing Style

You should use a consistent referencing style throughout your thesis. You can use any referencing style you like, or the one that the university is suggesting. However, you should use it consistently. If you are using a reference management software like Zotero (my prefered option), Mendeley, or EndNote you can easily change the style of your references.

You can also connect Zotero and Mendelely to your Overleaf account to use the references directly in your LaTeX document.

Placement of the References

People place the citations at different parts of the sentences and it is a bit of a personal writing style. I have two personal preferences depending on whether the citation is “numbered” or “named”.

Numbered

In numbered citations I prefer to place the citation at the end of the sentence. For example:

• Language is unique to humans [1].
• Computers can understand human language [2].
• The field of natural language processing is growing rapidly [3].

Named

I prefer placing the citation where a statement is made. For example:

• According to (Smith, 2020), language is unique to humans.
• Computers can understand human language (Jones, 2019).
• Doe (2018) showed that the field of natural language processing is growing rapidly.

These examples are not the only way to place the citations. You can choose your own style. However, you should be consistent in your choice.

Own your work

In my view, there is no problem in using assistive technologies (LLMs like ChatGPT and many others) for writing your thesis or analyzing your data, statistics, interpretations, or learning, and I even encourage you using them. However, after using such tools you should be able to judge the outcome. If the systems made outputs that included any sort of statement, you need to fact check them. Especially when it is used for content creation, you should have a complete understanding of the system’s output and it should not contain pieces that you do not understand or cannot explain. If you cannot explain a part of your thesis, it is a sign that you should not include it in your thesis.

Being my student, I ask you to include a comment in the LaTeX document of your thesis, where you use such tools, to indicate that you have used them. Examples:

% GPT-4o (grammar correction)
Today's Russian language has a limited number of accents due to the influence of the Soviet Union (Gorbachev, 1985). This has made the language easier to learn and understand for non-native speakers who wish to use within countries who use russian as one of their main languages.

I collected data from $60$ participants. 
% GPT-5 (Choice of statistical model)
I conducted a linear regression analysis to find the relationship between the age of the participants and their performance in learning verb forms in Amharic. The results for the regression analysis showed that the age of the participants is a significant predictor of their performance in learning verb forms in Amharic ($\beta = 0.45$, $p < 0.001$).
% Gemini Pro 1.5 (Interpretation of the results)
This result suggests that the older the participants are, the better they are in learning verb forms in Amharic. This is in line with the previous studies that showed that the age of the participants is a significant predictor of their performance in learning verb forms in Amharic.

In this post, we will focus on the public-key authentication method, which is more secure and convenient than using passwords. We will explain what SSH keys are, how they work, and how to create and use them on different operating systems, such as Windows 11, MacOS, and Ubuntu.

What are SSH keys?

SSH keys are two long strings of characters that are used to verify the identity of a user who wants to access a remote server. These keys are generated by the user on their local machine using a SSH utility. One key is private and stored on the user’s local machine. The other key is public and shared with the remote server or any other entity the user wishes to securely communicate with.

The private key is like a secret password that should never be revealed to anyone. The public key is like a lock that can be opened only by the matching private key. When the user tries to connect to the remote server, the server will check if the user has the correct private key that matches the public key stored on the server. If the keys match, the user is granted access. If not, the connection is denied.

SSH keys provide several advantages over passwords:

• They are more secure, as they are harder to guess or crack by brute force attacks.
• They are more convenient, as they do not require the user to remember or type a password every time they connect to the server.
• They can be used to automate tasks, such as running scripts or deploying applications, without human intervention.

How to create SSH keys?

To create SSH keys, you need to use a SSH utility that is available on your operating system. There are different tools and commands for different operating systems, but the basic steps are similar:

• Run a command to generate a pair of SSH keys, usually a 3072-bit or 4096-bit RSA key pair. You can optionally add a passphrase to further protect your private key.
• Save the keys in a secure location on your local machine, usually in a hidden folder named .ssh in your home directory. The private key will have no file extension, while the public key will have a .pub extension.
• Copy the public key to the remote server or any other entity you want to communicate with. You can use a utility called ssh-copy-id or manually append the public key to a file named authorized_keys in the .ssh folder on the remote server.

In the following sections, we will show you how to create SSH keys on Windows 11, MacOS, and Ubuntu, using different tools and commands.

Windows 11

Windows 11 comes with a built-in OpenSSH client that you can use to generate SSH keys. To use it, open the Windows Terminal or the Command Prompt and follow these steps:

• Type the following command to generate a pair of SSH keys and press Enter:

ssh-keygen -t rsa -b 4096 -C "your_email@example.com"

This will create a 4096-bit RSA key pair with your email address as a comment.

• You will be prompted to specify the file name and location for the keys. Press Enter to accept the default location, which is C:\Users\yourusername\.ssh\id_rsa for the private key and C:\Users\yourusername\.ssh\id_rsa.pub for the public key.

Enter file in which to save the key (C:\Users\yourusername/.ssh/id_rsa):

• You will be asked to enter a passphrase for the keys. This is optional, but recommended for extra security. Type a secure passphrase and press Enter. You will need to enter the same passphrase again to confirm it.

Enter passphrase (empty for no passphrase):
Enter same passphrase again:

• You will see a message that your keys have been generated, along with a fingerprint and a randomart image.

Your identification has been saved in C:\Users\yourusername/.ssh/id_rsa
Your public key has been saved in C:\Users\yourusername/.ssh/id_rsa.pub
The key fingerprint is:
SHA256:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx your_email@example.com
The key's randomart image is:
+---[RSA 4096]----+
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
+----[SHA256]-----+

• To copy the public key to the remote server, you can use the ssh-copy-id utility. Type the following command, replacing remote_username and server_ip_address with the username and IP address of the remote server. You will be prompted to enter the password of the remote user.

ssh-copy-id remote_username@server_ip_address

• Alternatively, you can manually copy the public key to the remote server. To do that, you need to open the public key file with a text editor, such as Notepad, and copy its contents. Then, you need to log in to the remote server using SSH, create a .ssh folder in your home directory if it does not exist, and append the public key to a file named authorized_keys in the .ssh folder. You can use the following commands to do that:

ssh remote_username@server_ip_address
mkdir -p ~/.ssh
echo 'paste_your_public_key_here' >> ~/.ssh/authorized_keys

MacOS

MacOS comes with a SSH utility called OpenSSH that you can use to generate SSH keys. To use it, open the Terminal application and follow these steps:

• Type the following command to generate a pair of SSH keys and press Enter:

ssh-keygen -t rsa -b 4096 -C "your_email@example.com"

This will create a 4096-bit RSA key pair with your email address as a comment.

• You will be prompted to specify the file name and location for the keys. Press Enter to accept the default location, which is /Users/yourusername/.ssh/id_rsa for the private key and /Users/yourusername/.ssh/id_rsa.pub for the public key.

Enter file in which to save the key (/Users/yourusername/.ssh/id_rsa):

• You will be asked to enter a passphrase for the keys. This is optional, but recommended for extra security. Type a secure passphrase and press Enter. You will need to enter the same passphrase again to confirm it.

Enter passphrase (empty for no passphrase):
Enter same passphrase again:

• You will see a message that your keys have been generated, along with a fingerprint and a randomart image.

Your identification has been saved in /Users/yourusername/.ssh/id_rsa
Your public key has been saved in /Users/yourusername/.ssh/id_rsa.pub
The key fingerprint is:
SHA256:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx your_email@example.com
The key's randomart image is:
+---[RSA 4096]----+
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
+----[SHA256]-----+

• To copy the public key to the remote server, you can use the ssh-copy-id utility. Type the following command, replacing remote_username and server_ip_address with the username and IP address of the remote server. You will be prompted to enter the password of the remote user.

ssh-copy-id remote_username@server_ip_address

• Alternatively, you can manually copy the public key to the remote server. To do that, you need to open the public key file with a text editor, such as TextEdit, and copy its contents. Then, you need to log in to the remote server using SSH, create a .ssh folder in your home directory if it does not exist, and append the public key to a file named authorized_keys in the .ssh folder. You can use the following commands to do that:

ssh remote_username@server_ip_address
mkdir -p ~/.ssh
echo 'paste_your_public_key_here' >> ~/.ssh/authorized_keys

Ubuntu

Ubuntu comes with a SSH utility called OpenSSH that you can use to generate SSH keys. To use it, open the Terminal application and follow these steps:

• Type the following command to generate a pair of SSH keys and press Enter:

ssh-keygen -t rsa -b 4096 -C "your_email@example.com"

This will create a 4096-bit RSA key pair with your email address as a comment.

• You will be prompted to specify the file name and location for the keys. Press Enter to accept the default location, which is /home/yourusername/.ssh/id_rsa for the private key and /home/yourusername/.ssh/id_rsa.pub for the public key.

Enter file in which to save the key (/home/yourusername/.ssh/id_rsa):

• You will be asked to enter a passphrase for the keys. This is optional, but recommended for extra security. Type a secure passphrase and press Enter. You will need to enter the same passphrase again to confirm it.

Enter passphrase (empty for no passphrase):
Enter same passphrase again:

• You will see a message that your keys have been generated, along with a fingerprint and a randomart image.

Your identification has been saved in /home/yourusername/.ssh/id_rsa

OK, I will continue generating the blog post.

• You will see a message that your keys have been generated, along with a fingerprint and a randomart image.

Your identification has been saved in /home/yourusername/.ssh/id_rsa
Your public key has been saved in /home/yourusername/.ssh/id_rsa.pub
The key fingerprint is:
SHA256:xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx your_email@example.com
The key's randomart image is:
+---[RSA 4096]----+
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
|                 |
+----[SHA256]-----+

• To copy the public key to the remote server, you can use the ssh-copy-id utility. Type the following command, replacing remote_username and server_ip_address with the username and IP address of the remote server. You will be prompted to enter the password of the remote user.

ssh-copy-id remote_username@server_ip_address

• Alternatively, you can manually copy the public key to the remote server. To do that, you need to open the public key file with a text editor, such as Gedit, and copy its contents. Then, you need to log in to the remote server using SSH, create a .ssh folder in your home directory if it does not exist, and append the public key to a file named authorized_keys in the .ssh folder. You can use the following commands to do that:

ssh remote_username@server_ip_address
mkdir -p ~/.ssh
echo 'paste_your_public_key_here' >> ~/.ssh/authorized_keys

How to use SSH keys?

Once you have created and copied your SSH keys, you can use them to connect to the remote server without entering a password. To do that, you need to use the ssh command with the -i option to specify the path to your private key. For example, to connect to the remote server using the private key stored in the default location, you can use the following command:

ssh -i ~/.ssh/id_rsa remote_username@server_ip_address

If you have added a passphrase to your private key, you will be asked to enter it once per session. If you want to avoid typing the passphrase every time, you can use a utility called ssh-agent to store the passphrase in memory. To do that, you need to run the following commands:

eval $(ssh-agent)
ssh-add ~/.ssh/id_rsa

You will be prompted to enter the passphrase once, and then it will be cached by the ssh-agent. You can then connect to the remote server using the ssh command without entering the passphrase again.

Conclusion

In this post, we have learned what SSH and SSH keys are, how they work, and how to create and use them on different operating systems. SSH keys are a secure and convenient way to access and communicate with remote servers over the internet. We hope you have found this post useful and informative. If you have any questions or feedback, please leave a comment below. Thank you for reading! 😊

Communication

Slack

I preffer to use Slack for work communication. I need to invite you to my workspace using Slack Connect. I will let you know about it.

Materials

User Interfance/Experience (UI/UX) Design

Go through this part if your project includes a user interface design. If not, skip to the next section.

Prototyping

Figma (Recommended)

1. Create an account on Figma
2. Register for Figma Education using your KTH email

Version Control

Needless to say that we will need a version control for our programming materials

GitHub

1. Create an account on GitHub if you do not have one already
2. Register for GitHub Education and get verified as a student so that you can access the premium features for free and so many other benefits that I encourage you to use, such as GitHub Copilot
3. Activate your GitHub Copilot
4. Create a private repository for your degree project/thesis and add me as a collaborator using my GitHub username @horotat

Open Science Framework (OSF)

A healthy research ethics means not to be worried about sharing your work with others. If we are honest in all what we do, then we sholdn’t worry about sharing the details of our analsys, data, and methods with others. An open science attitude is a good way to give the other researchers the opportunity to reproduce your work and build on it. OSF is a platform that facilitates this process. If you are going to work on a scientific project with me, I recommend you to:

1. Create an account on OSF
2. Create a project for your degree project/thesis
3. Add me as a contributor using my KTH email
4. Keep the repository private for the time being
5. In our first meeting we will put the proposal on the OSF repo and build on top of that

Time Management

I am not strict about your preferred way of time management, but it is important that you have a framework and we follow up with the agreed method. You need to take care of the time management yourself but present your plan and updates, each time we meet. You can as well just use Slack Canvas or written text. But if you want some flaivour, here are some recommendations of the systems that I personally have used and liked:

• (Recommended) Notion(With free student subscription).
• (Recommended) Microsoft To Do
• Other platforms:
  • ClickUp
  • Trello
  • Todoist
  • Google Tasks

Literature

Search & Review

During your project with me, you will learn about how to search for literature and find relevant studies to shape your project in a proper academic way. We will discuss the techniques all together but before hand you can familiarize yourself with the following platforms:

If you are KTH student, review the KTH Library information about Literature search.

Here is a list of platforms to use for literature search and review:

• Search
  • Google Scholar
  • Scopus
  • Web of Science
• Evaluate
  • Elicit
  • Scite

Reference Management

You will need a way to keep track of the references and literature that you find so that later you can easily cite them. There are some softwares that can do that. You can find information about many of them on this page of KTH Library. My preffered software is Zotero.

Zotero

1. Install Zotero on your computer and browser
2. Create an account on Zotero
3. Create a group for your degree project/thesis, and add me as a member using my KTH email

Writing

LaTeX, on Overleaf

I recommend you to use LaTeX for writing your thesis and personally I am most comfortable with LaTeX for collaboration and commenting on your writings. For this purpose follow these steps:

1. Go to Overleaf. Sign in with your KTH account (SSO)
2. Create a project and add me as a collaborator using my KTH email
3. Go to the setting of your profile and connect your Overleaf account to your GitHub and Zotero accounts (included in the Overleaf premium with your KTH account)

Overleaf is pretty straight forward, but if you do not know LaTeX, the pages below provide very nice and quick tutorials:

• Learn LaTeX in 30 minutes: Quick option
• Learn LaTeX: More comprehensive option

Apart from these there is a ton of resources online. Just google it!

Facilities for My Students

As long as you work with me I could try to provide you with the following facilities:

Computational Resources

I can give you SSH access to my server for the duration of your project to exclusively be used for your processes. The server configuration is as follows:

| OS | Ubuntu 22.04 LTS | | CPU | Intel(R) Core(TM) i9-10850K @ 3.60GHz | | GPU | GeForce RTX 3090, 24GB | | RAM | DDR4 128GB | | Storage | 1TB SSD, 4TB HDD |

If you need heavy computation in your project send me your SSH public key, mentioning your needs and demands, and I will give you access to the server.

If you are not familiar with SSH, you can find a tutorial in this post.

Office Space and Access to TMH

If you need it, I can provide you with office space within TMH division. This is not promised and depends on the availability of the space.

To get access to TMH facilities, send me your Personal Number (Personnummer) (or birthdate if you do not have a Swedish personal number.).

Computer

Depending on what you need to do, I might be able to get you a laptop for the duration of your project. This is not promised and depends on the availability of the laptops.