FIRST EDITION
ARC Summer Edition 2026
|
1 June –
1 September 2026
FEAA · UAIC
Iași, Romania
|
The ARC Summer Edition 2026
brought together a selected cohort of students for a summer of practical work
in Artificial Intelligence, data analysis, scientific literature, and research methods.
The edition combined structured workshops with research onboarding,
individual assignments, technical notebooks, literature-search exercises,
dataset discovery, and group discussion. Participants were introduced to AI
and data-analysis tools while also learning how a broad area of interest can
become a research topic, how literature and datasets are identified, and how
an analytical idea begins to take shape.
The Summer Edition also served as the first structured onboarding experience
for the ARC community. Members worked together, chose individual themes,
completed research-oriented homework, and gradually moved from introductory
exercises toward more independent work.
The first Summer Edition received
52 applications.
Twenty-eight applicants were initially selected for the restricted cohort,
and 20 members completed the edition.
52
applications received
|
28
applicants selected
|
20
members completed the edition
|
71%
completion among accepted members
|
Four main workshops provided the structured backbone of the Summer Edition,
moving from introductory Machine Learning concepts toward exploratory data
analysis and scientific-literature research.
WORKSHOP 01 · 4 HOURS
Introduction to AI & Machine Learning
An introduction to Artificial Intelligence and Machine Learning,
common learning tasks, datasets, model development, and the basic
structure of an applied ML workflow.
|
WORKSHOP 02 · 4 HOURS
Introduction to Exploratory Data Analysis
Practical work with dataset structure, distributions, missing values,
descriptive statistics, visualization, and initial data-quality assessment.
|
WORKSHOP 03 · 4 HOURS
Introduction to Scientometrics
Scientific-literature data, bibliographic databases, search strategies,
publication metadata, citation analysis, and literature-based research workflows.
|
WORKSHOP 04 · 3 HOURS
Advanced Exploratory Data Analysis
A deeper look at data relationships, distributions, anomalies,
feature behaviour, comparative analysis, and more structured
approaches to understanding a dataset before modelling.
|
A large part of the Summer Edition focused on what happens before an experiment
or model exists. Participants worked through the early decisions involved in
starting a research project: selecting a topic, understanding the literature,
finding data, and deciding what can realistically be investigated.
Choosing an individual research direction
Members selected their own themes and completed guided homework
intended to move from a broad area of interest toward a clearer
topic and potential research question.
|
Searching the scientific literature
Participants constructed search strings for Scopus and Web of Science
and discussed how terminology, database coverage, and query design
influence what literature is retrieved.
|
Understanding technical papers
Members learned to separate
survey papers, dataset papers, benchmark papers, and model papers
and to consider what each type can contribute to a research project.
|
Finding suitable datasets
Participants searched for datasets connected to their chosen themes,
examined available documentation, and discussed whether the data
supported the type of question they wanted to investigate.
|
|
04
|
Reading research critically
|
|
Members were asked to read a published review paper before a dedicated
discussion session. The paper was then examined collectively in class,
with attention to its research question, literature-selection process,
structure, methods, claims, strengths, and limitations.
The objective was to move away from reading a paper simply as something
that has already been “approved” by publication. Participants were encouraged
to question methodological decisions, identify what the authors had actually
established, and consider what they would have done differently.
Five practical notebooks introduced common data modalities and the first
steps involved in working with unfamiliar datasets.
Across the exercises, members practiced finding datasets on Kaggle, importing
them into a working environment, inspecting their structure and metadata,
identifying cleaning requirements, and preparing the data for exploratory analysis.
|
Tabular data
|
Time series
|
Sound
|
Images
|
Text
|
ARC on Kaggle
Participants were introduced to Kaggle for dataset discovery,
notebooks, experimentation, and sharing work within a technical community.
ARC now maintains its own Kaggle community for continued collaboration
beyond the Summer Edition.
ARC – The AI Research Club on Kaggle ↗
Community and confidence
Not every Summer Edition activity was a formal workshop.
Some meetings were less structured and focused on building confidence,
helping members get to know one another, and becoming comfortable
with the tools and habits used throughout the club.
These sessions included discussions about prompting and contemporary
AI tools, Kaggle onboarding, notebook use, dataset discovery,
individual research-theme selection, guided searches for technical
papers, research homework, and group discussion around members'
ideas and problems.
|
Workshop materials
Workshop notebooks, exercises, internal resources, and supporting
materials remain available to ARC members only.
This public page documents what was covered. The learning materials
remain part of the members' working environment.
|
|
07
|
Summer Edition members
|
|
The following members completed the ARC Summer Edition 2026.
|
08
|
After the Summer Edition
|
|
The Summer Edition was designed as a starting point, not a one-off course.
Members who completed the programme can continue through ARC's recurring
activities, workshops, research discussions, project work, collaborative
initiatives, and future events.
The experience also helped establish the structure that ARC now uses for
research onboarding: practical work, scientific-literature literacy,
individual topic development, and an expectation that members actively
test, question, and discuss what they are learning.
