The biggest European conference about ML, AI and Deep Learning applications
running in person in Prague and online.

Machine Learning Prague 2024

In cooperation with

, 2024


World class expertise and practical content packed in 3 days!

You can look forward to an excellent lineup of 40 international experts in ML and AI business and academic applications at ML Prague 2024. They will present advanced practical talks, hands-on workshops and other forms of interactive content to you.

What to expect

  • 1000+ Attendees
  • 3 Days
  • 40 Speakers
  • 10 Workshops

Phenomenal Speakers

Practical & Inspiring Program


O2 Universum, Českomoravská 2345/17a, 190 00, Praha (workshops won't be streamed)


Room D2 Room D3 Room D4 Room D6 Room D7
coffee break

Using Graph Neural Networks to improve telecommunication networks

Room D2

Massimiliano Kuck, Sopra Steria SE
Malte Wagner, Sopra Steria SE

This workshop aims to provide an in-depth understanding of modeling and optimizing telecommunication networks using Graph Neural Networks (GNNs). Telecommunication networks characterized by a complex structure of interconnected nodes can be naturally modelled as graph data structures. However the utilization of graph data structures in Machine Learning poses unique challenges. The workshop focuses on various aspects of integrating GNNs into the telecommunication networks to enhance their performance and customer satisfaction. The implementation of GNNs introduces a new level of sophistication that allows intelligent decision making effective network mapping prediction and optimization. The key component of the workshop is a practical use case where participants will work with a Neo4J graph database. This database aids in visualizing and analyzing the complex connections in the network to provide a base for the GNN. With the use of GNNs valuable insights and predictions on network activity can be made efficiently. They leverage the rich information present in the network structure. Using these predictions to optimize network configuration settings and address network problems could potentially result in lower operational costs improved network efficiency and higher customer satisfaction. The workshop aims for participants to learn the practical application of modeling telco networks as graph structures implementing GNNs and interpreting their predictions for optimization. Moreover it's noteworthy that graph structures are not confined to telecommunications they can be found spanning across any industry sector and GNNs have versatile applicability in different use cases. It is an excellent opportunity for practitioners researchers and anyone interested in graph data modeling and machine learning to learn from industry experts. By the end of this workshop participants will have a more profound understanding of the potential of GNNs enabling them to apply this technology to optimize various network systems.

Real-Time Anomaly Detection in Python

Room D3

Tomáš Neubauer, Quix
Daniil Gusev, Quix

Get to grips with real-time anomaly detection in Python by working through a use case for detecting cyclist crashes. In this hands-on workshop we will learn how to build a streaming data pipeline using Kafka to handle telemetry events from a bicycle sensor/fitness app. From here we will collect data to label and train an ML model and deploy it to predict crashes as they happen in real time.

Chatting with your Data: A Hands-on Introduction to LangChain.

Room D4

Marina Volkova, HumbleBuildings

In this interactive workshop we are covering how to handle models to work with your own data whether it's text PDFs CSVs or SQL database. We'll show you how to efficiently run Language Models (LLMs) on a standard 16GB RAM machine. During the workshop you will gain hands-on experience with the LangChain framework and have the opportunity to create an application that allows you to interact with your data through natural language. The beauty of this workshop lies in its accessibility as we leverage low-code frameworks that make it suitable for a broad audience. All you need is a laptop with an internet connection access to the Google Colab environment and optionally an OpenAI API key to achieve even more impressive results. Join us on this exciting journey of data interaction and exploration!

Empowering Question Answering Systems with RAG: A Deep Dive into Retrieval Augmented Generation

Room D6

Peter Krejzl, Emplifi
Jan Rus, Emplifi

Join us for a workshop on Retrieval Augmented Generation (RAG) a groundbreaking approach in machine learning that seamlessly integrates information retrieval with text generation. Dive deep into the construction of advanced question-answering systems that leverage private knowledge bases moving beyond mere document retrieval to generate coherent and natural language responses. Throughout the workshop participants will benefit from hands-on experiences utilizing real datasets and the latest LLM techniques ensuring practical comprehension. By harnessing the power of semantic search and private databases RAG-based system promises a new user experience and proprietary context-specific solutions. - Introduction to RAG: We will begin by introducing the concept of Retrieval Augmented Generation (RAG) and its importance in the landscape of modern machine learning solutions. - Dual Strength: RAG uniquely combines an information retrieval component with a text generation model offering a two-pronged approach to problem-solving. - Private Knowledge Bases: Traditional systems rely heavily on public datasets. In this workshop participants will learn to build systems that leverage private knowledge bases ensuring proprietary and context-specific responses. - Beyond Simple Retrievals: It's not just about finding the right documents. Our RAG-based system will not merely retrieve relevant articles but will craft answers in coherent and natural language enhancing user experience. - Practical Implementation: We will guide attendees through the process of building a question-answering system powered by semantic search and the latest LLM (Language Model) techniques. - Hands-on Experience: Participants will get hands-on experience working with real datasets and observing live demonstrations ensuring a practical understanding of the RAG system. - Future Potential: The workshop will conclude with a discussion on the future potential and advancements of RAG in diverse applications. Join us for the workshop and elevate your understanding of how RAG is reshaping the frontier of question-answering systems.

Automated Evaluation of LLM based systems

Room D7

Marek Matiáš, O2/Dataclair
Ondřej Finke, O2/Dataclair
Alexandr Vendl, O2/Dataclair

Development of complex LLM based solutions cannot be done without robust system for evaluation of the outputs. Whenever you make changes to models prompts or other components of the system you need a metric to find out whether the performance improved overall. We will outline possible solution to this problem including broader picture as well as hands-on exercise.

coffee break

Building OpenAI Plugins: Deep Dive into Microsoft Semantic Kernel (SK)

Room D2

Daniel Costea, European Agency

Microsoft Semantic Kernel (SK) is a new technology that enables the integration of AI Large Language Models (LLMs) like GPT-3.5-Turbo GPT-4 and DALL-E 3 from OpenAI or Azure OpenAI with conventional programming languages like C# Python and Java. SK brings together several key components to provide planning and execution capabilities. These components include a robust kernel that provides the foundation for all other components plugins (formerly known as skills) for performing specific tasks connectors for interfacing with external systems memories for storing information about past events steps for defining individual actions and pipelines for organizing complex multi-stage plans. In this hands-on workshop we explore how to build various plugins for: - building a semantic interface for an existing API using plugins and execution plans containing semantic and native functions. - building a GPT-powered chat enriched by real-time information and memories enhanced through RAG (Retrieval-Augmented Generation) capabilities. - building a cutting-edge generative model using DALL-E 3 and multi-modal input.

Finetuning Open-Source LLMs to small languages

Room D3

Petr Simecek, Mediaboard
David Jiříček, Mediaboard
Filip Sedlák, Mediaboard

Large Language Models (LLMs) represent a remarkable advancement in artificial intelligence (AI) boasting the capability to generate and comprehend human language. Derived from extensive training on vast text and code datasets these models excel in a variety of tasks such as translation summarization and question answering. However a major limitation arises when these LLMs predominantly trained on English data are applied to other languages particularly smaller languages like Czech. Notable models like ChatGPT Bard and Claude exhibit proficiency in Czech with minimal grammatical and stylistic errors. Yet many contemporary open-source LLMs influenced heavily by English-centric datasets fail to address even basic Czech queries.So what are the choices? At Monitora our initial experiments with ChatGPT for text summarization have now transitioned to Llama2 7B primarily due to privacy considerations. We are also evaluating Mistral models introduced in September. Within this workshop I will give introduction into QLoRA adapters and demonstrate that instruction finetuning of such models is possible even with limited resources (a single consumer GPU). We will see that models originally speaking a very broken language improve significantly by this process. In addition to the technical insights I envision this workshop as a collaborative forum. Rather than just being a traditional presentation it aims to be a platform for knowledge exchange. Attendees are encouraged to contribute their insights plans or experiences related to the application of LLMs in small languages. For structure I would like to limit each interested participant to 5 slides and a presentation time of 5 minutes.

Unlocking the Power of Active Learning: A Hands-on Exploration

Room D4

Fabian Kovac, St. Pölten University of Applied Sciences
Oliver Eigner, St. Pölten University of Applied Sciences

In today's ever-evolving landscape of Artificial Intelligence and Machine Learning staying at the cutting edge is not just an advantage it's a necessity. Active Learning has emerged as a powerful technique that has the potential to revolutionize how we train machine learning models. With this hands-on workshop we are providing attendees with a comprehensive understanding of its relevance benefits and practical applications. Active Learning is a paradigm-shifting approach that focuses on enabling machines to learn more efficiently from limited labeled data by actively selecting the most informative examples for annotation. It plays a pivotal role in various industries and research domains offering solutions to some of the most pressing challenges in AI and machine learning. By actively involving human experts in the loop Active Learning not only reduces annotation costs and efforts but also accelerates model development making it particularly relevant for resource-constrained environments where only limited labeled data is available. This workshop will be a hands-on immersive experience providing a solid foundation of the theoretical underpinnings ensuring attendees grasp the core concepts. Participants will have the opportunity to apply Active Learning techniques to real datasets gain practical experience in selecting informative data points training models and observing the impact on model performance. Furthermore we will share best practices and pitfalls to avoid when implementing Active Learning in real applications. This workshop promises to equip attendees with the knowledge and skills they need to harness the full potential of Active Learning in their research and industry applications. By the end of the workshop participants will be well-prepared to incorporate this cutting-edge technique into their AI and Machine Learning endeavors accelerating progress and achieving superior results.

Practical Tips for Deep Transfer Learning

Room D6

Yauhen Babakhin, H2O
Martin Barus, PureStorage

In this workshop we will start by introducing Transfer Learning and its benefits in Deep Learning. Then we will have an interactive session discussing and implementing best practices for tuning such models and improving their results. The workshop will be accompanied by an in-class Kaggle competition in image classification which will have a ready-to-use code baseline in PyTorch. Participants will be able to directly apply the concepts we discuss to real data. Additionally they will have the chance to introduce their own improvements to the solution baseline and share their findings afterwards. 

Power of Physics-ML, a hands-on workshop with open-source tools

Room D7

Idrees Muhammad, Turing Artificial Intelligence

In this workshop participants will learn how to combine the power of neural networks with the laws of physics to solve complex scientific and engineering problems using open-source tools. Physics-Informed Neural Networks (PINNs) have gained popularity in various fields including fluid dynamics material science and structural engineering for their ability to incorporate physical principles into machine learning models. Attendees will learn how to leverage open-source tools to build and train PINNs enabling them to model and solve complex physical systems efficiently.


O2 Universum, Českomoravská 2345/17a, 190 00, Praha (and on-line)

Registration from 9:00

Welcome to ML Prague 2024

LLMs, Reasoning, and the Path to Intelligence

Murray Campbell, IBM T. J. Watson Research Center

LLMs have shown impressive performance across a wide range of tasks, including tasks that require reasoning.  Despite these results, there is considerable evidence that the reasoning capabilities of LLMs have serious limitations.  Given the significant benefits of reasoning, including improved generalization, ability to incorporate new knowledge, better explainability and reliability, it is important to better understand what LLMs can and cannot be expected to do.  In this talk I will provide an overview of reasoning and its benefits, characterize some of the reasoning limitations of LLMS, and discuss approaches to move toward AI systems that can reason more effectively.

Enhancing Semantic Search: A Case Study on Fine-tuning with Noisy Data

Barbora Rišová,

The efficient semantic search plays a pivotal role in the search engine, with the vector space method proving to be a fast and effective means of mining relevant documents for user queries. However, the effectiveness of this approach is contingent upon the alignment of the vector space with the specific requirements of the search engine. This presentation delves into the realm of contrastive learning, an efficient technique for acquiring optimal vector representations of diverse modalities, including text, images, or both. While contrastive learning offers a robust foundation, the preferential approach for downstream tasks involves fine-tuning models on labeled data. Our talk will shed light on the innovative application of contrastive learning within the fine-tuning phase of semantic models, leveraging noisy “click” user data to augment the search engine's capabilities. We will present compelling real-world use cases spanning various modalities within the search domain, including images, products, web documents, and advertisements.

Quantum-inspire and Quantum Machine Learning

Alexander Del Toro Barba, Google

Quantum-inspired and Quantum Machine Learning offer the potential to solve problems intractable for classical computing. This presentation delves into the advantages of these approaches, where quantum computing can accelerate traditional machine learning methods, where traditional machine learning methods provide an advantage, and with a look at potential applications.


Poster session:

  • Accurate Answering of Questions Based on Existing Scientific Literature Implemented on a Shoestring (Tomáš Kubeš)
  • AI-powered system for automatic processing of medical notes (Aleksandra Kowalczuk)
  • ML platform for Low Latency Processing of Big Data (Radim Špigel, Tomáš Sedláček, Vít Líbal)
  • Optimizing the Geothermal Drilling Process Using Artificial Intelligence Methods (Henning Knauer, Shahin Jamali, Kevin Mallin)
  • Harnessing Evolution: The Pathway to Optimizal Quantum Circuit (Libuše Horáčková, Ivan Zelinka)
  • Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMs (Patricia Schmidtova, Simone Balloccu, Mateusz Lango, Ondřej Dušek)
  • Metal Price Prediction on the London Metal Exchange: A Deep Learning Approach (Jaroslav Kovalenko, Lukáš Hefka)

Supercharging Recommendation Systems with Large Language Models

Amey Dharwadker, Meta

Recommender systems have evolved into an indispensable part of our daily experiences, delivering tailored recommendations across diverse applications that cater to individual preferences. Yet, traditional recommendation models often struggle with the complexities of comprehensively understanding user interests and adapting to the ever-expanding landscape of recommendation scenarios. Enter Large Language Models (LLMs), renowned for their unparalleled strength in language comprehension, generation, generalization and reasoning. In this talk, we will explore the transformative influence of LLMs on recommender systems. We'll provide a comprehensive overview of the strategies for seamlessly integrating LLMs into recommendation pipelines, shedding light on where and how these integrate effortlessly into the recommendation ecosystem while addressing practical challenges and opportunities that emerge. We will also delve into methodologies such as fine-tuning and prompt tuning, showcasing how these techniques unlock the full potential of LLMs. Finally, we will lay out a roadmap of exciting opportunities, offering a glimpse into a future where recommendations are not only highly personalized but also exceptionally efficient.

Practical LLM Fine-Tuning For Semantic Search

Roman Grebennikov, Delivery Hero SE


Have you ever wondered if the typical semantic search recipe of “embed all your documents and do the vector search” holds up in practice? Will it fit a narrow medical, law, or hardware domain? Can it handle non-English languages well? Taking a one-size-fits-all model from Huggingface does not always do the trick for your specific business needs. But the good news is that you can fine-tune such models to better fit your world, and open-source tools like sentence-transformers and nixietune are here to help you! In this talk, we will take a hands-on journey to fine-tune your own embedding model for semantic search. What kind and how much data do you need, how do you train a bi-encoder and a cross-encoder, and how far can you go quality-wise with a single GPU? If you're ready to make semantic search work for you, come join us!

Perspective Taking in Large Language Models

Lucie Flek, University of Bonn

While LLMs outperform humans in an ever-broadening range of tasks, they remain far behind in the ability to explain somebody else’s feelings, thoughs and behavioral drivers. This skill requires perspective-taking, the process of conceptualizing the point of view of another person. Something that humans do effortlessly and instinctively all the time - tracking the concealed mental states of others, encompassing knowledge, intentions, beliefs, and desires. This ability holds a central role in various aspects of human existence, including social interactions, self-consciousness, and ability to empathize. Foreseeing the actions or reactions of others is also for LLMs a key to choose the best action to take next. Enhancing perspective-taking capabilities of LLMs can unlock their potential to react better and safer to hints of sadness, anger, or distress, to recognize sarcasm, to engage in a more receptive argumentation, or to target an explanation to an audience. In this talk, I will review recent approaches to modelling the human context in transformers holistically -  including personal history, beliefs, and social environment – and their effect on downstream NLP tasks. I will present our recent perspectivism-motivated experiments, and discuss further opportunities for bringing the human-centered paradigm into LLMs.


Application of Machine Learning in Metagenomics

Enes Deumić, Genome Institute of Singapore

Genomics is a rapidly expanding field that has recently benefited from a convergence of numerous technological and scientific advancements. In genomics, we confront a unique set of challenges, such as intricate data pipelines and metrics, alongside evaluation methods that are still not well established. In this presentation we will delve deeper into the field of metagenomics and how we use machine learning to classify microorganisms from free samples. We will show how to combine transformer models and hashing models for maximal performance.

Translating Mobile Network Signals to Roads with Transformers

Stefan Josef, Dataclair

In recent years, Transformers and self-supervised learning have achieved unprecedented success. This is exemplified by models like ChatGPT, which have garnered significant attention even among the general public. While numerous pre-trained language models are readily available for text data, either as commercial APIs or open-source models, many companies deal with large volumes of tabular, time-series, or graph data. This necessitates the creation and training of custom models for many diverse use cases. Map matching refers to the procedure of converting sparse and noisy signals into precise positions on a road network. Drawing inspiration from the methods that propelled success in the text domain, such as language modeling, machine translation, and transfer learning, Dataclair, AI centre at O2 Czech Republic, has trained its own sequence-to-sequence Transformer models, which effectively translate mobile network signals into specific road segments. By providing more accurate aggregated and anonymized data, telecommunication companies can offer valuable insights to public decision-makers, for example aiding in enhanced traffic planning.

Developing and Up-Scaling RNA-Based Vaccines and Therapeutics Production Using Big Data and Generative AI

Andreea Mihailescu, Johns Hopkins University

Beyond its potential for infectious disease prophylaxis, RNA technology has emerged as one of the most promising techniques for the rapid development of treatments envisaging a wide range of pathologies ranging from cancer to cardiovascular, autoimmune, or rare diseases. While conventional vaccines and pharmaceuticals rely on the active use of cell cultures, RNA manufacturing is based on a relatively simple, scalable, and affordable cell-free production line. Furthermore, RNA-based technology is known to hold the major advantage of being disease-agnostic, hence exhibiting high versatility in terms of therapeutic scope. Nevertheless, a decisive factor in the successful design of such innovative therapeutics for personalized medicine is the use of big data and AI technologies. These account not only for the fast RNA sequencing and proteomics data analysis but also for the identification and selection of viable and relevant RNA codes with low immunogenicity, maximum potency, and superior production yields. Multi-omics data integration helps molecule design and discovery, ensuring drug connectivity analysis and drug response prediction. This presentation will showcase how to harness the potential of a disease-agnostic RNA platform for personalized medicine by making use of custom-designed big data architecture, generative AI, and deep learning but also biotech digital twins and software sensors, all within the framework of Quality by Digital Design (QbDD) with fast and interactive visualizations provided by the solution’s designated dashboard. All these have led to the development and up-scaling of the existing production processes both in terms of volume as well as timelines at unprecedented speed.


Building Responsible and Safe Generative AI Applications

Mehrnoosh Sameki, Microsoft

As large language models (LLMs) become more widely adopted, it is crucial to understand their effective utilization, copilot development, evaluation, operationalization, and monitoring in real-world applications. This session will provide insights into incorporating responsible AI practices and safety features into your generative AI applications. You will gain knowledge on assessing your copilots and generative AI applications, mitigating content-related risks, addressing hallucinations, jailbreak, and copywrite issues, ensuring fairness, and enhancing the overall quality and safety of your copilot.

Protecting Privacy with AI During Testing of Automated Cars

Mateus Riva, Valeo

Development of smart/autonomous cars relies on testing in the real world, with real traffic. During this type of testing, thousands of hours of camera footage is recorded and used for learning, improving and testing the autonomous mode for the cars. We have developed and studied several algorithms that can find human faces and potentially readable license plates in every frame of a video and anonymize it. Why do we anonymize? When is it mandatory? How much of it is necessary?

Networking & drinks

Conference day 1

O2 Universum, Českomoravská 2345/17a, 190 00, Praha (and on-line)

Doors open at 08:30

Deep Learning Discovery of New Exoplanets

Hamed Valizadegan, NASA

The Kepler and TESS missions have yielded an astounding 100,000 potential transit signals, paving the way for an intricate process of distillation to identify viable exoplanet candidates. In response to this formidable challenge, we introduced ExoMiner, a groundbreaking deep neural network meticulously designed for the classification of transit signals in the search for exoplanets. ExoMiner played a pivotal role in validating and authenticating 301 previously undiscovered exoplanets. This keynote voyage commences with a sweeping overview of the captivating realm of exoplanetary exploration. As we delve into the heart of our presentation, we embark on an exploration of the distinctive attributes that set ExoMiner apart, unraveling the intricate web of factors that contribute to its remarkable accuracy. Our narrative extends beyond the confines of ExoMiner as we navigate through the integration of additional machine learning models seamlessly layered atop the ExoMiner architecture using basic domain knowledge information. This collaborative synthesis has already led to the identification of an additional 69 exoplanets, showcasing our unwavering commitment to pushing the boundaries of exoplanetary discovery. The journey of implementing machine learning in domains of such complexity is riddled with practical challenges. Guaranteeing consistent performance and earning the trust of domain experts in the reliability of results becomes an arduous task. This keynote will intricately delve into the practical intricacies, providing profound insights into the delicate balance between machine intelligence and the discerning eye of scientific scrutiny.

Small-Data Deep Learning and Its Applications to Diagnostic Aid and Virtual AI Imaging

Kenji Suzuki, Tokyo Institute of Technology

Deep leaning became one of the most active areas of research in many fields including artificial intelligence (AI), computer vision, autonomous driving, factory automation, robotics, and medicine. Deep learning is a breakthrough technology in the machine learning field, as it exceeded human performance when it was trained with a large number of cases, "big data". In many areas such as medical image diagnosis and low-volume productions in factories, however, collecting and annotating a large number of cases of patients or products is a big challenge. Deep learning models require 10,000 to 100,000 cases of patients or factory parts to adequately train, which would take months or years to collect. In the past research, less attentions were given to deep learning models that did not require big data. In this talk, "small-data" deep learning models that can be trained with a limited number of cases is introduced. Small-data deep learning is defined as deep learning models that offer the performance equivalent to the "big-data" deep-learning models but require a small number of cases (< 100) to adequately train, thus, reducing the number of necessary cases by a factor of 100-1,000. My group has been actively studying on deep learning in medical imaging in the past 25 years, including ones of the earliest deep-learning models for medical image processing, semantic segmentation of lesions and organs, lesion/organ enhancement, and detection and classification of lesions in medical imaging. We investigated the development of novel small-data deep-learning models and their applications to AI-aided diagnostic systems (“AI doctor”) and deep-learning-based imaging for medical diagnosis (“AI medical imaging”). We investigated a required number of cases for our small-data deep-learning model for AI-aided analysis of lesions in medical images. We demonstrated that our small-data deep-learning model, called a massive-training artificial neural network (MTANN), was able to achieve the state-of-the-art performance with a small number of cases (> 100). In the application to liver tumor segmentation, our small-data deep-learning model achieved the performance equivalent to the world-competition-winning deep-learning model with a very small required number of training cases of only 14. Our other verified studies include 1) AI systems for cancer detection, analysis, and diagnosis with medical images, and 2) virtual AI imaging systems for separation of bones from soft tissue in chest radiographs and those for radiation dose reduction in CT and mammography. Thus, small-data deep-learning models that can be trained with a small number of cases would be the next breakthrough technology in the AI field, and it would fill in the gap between "big-data" areas and "small-data" areas where many cases are not available such as in medicine, healthcare, and low-volume production industries.

Instant Insight: The Rise of Real-Time Online ML

Madalina Ciortan,

In the rapidly evolving landscape of the travel industry, the ability to adapt and respond in real-time is not just an advantage, but a necessity. This talk offers an insightful overview of real-time online machine learning (ML), a pivotal area in modern data science. We will explore the core principles of online ML, contrasting it with traditional batch learning, and highlighting its significance in processing continuous data streams. Key focus areas include a review of various online learning algorithms, their adaptability to evolving data, and the balance between responsiveness and stability. The presentation will examine both the advantages, such as real-time data handling and adaptability, and the challenges, including computational demands and potential overfitting, associated with online ML methods. Through select examples, we will demonstrate the practical applications and limitations of these techniques across different sectors. Concluding with a look at future trends in online ML, this talk aims to provide a concise yet comprehensive understanding of the field, enabling participants to appreciate its impact and potential in diverse applications.


Churn Detection and Explanation for a Modern Taxi-hailing Company

Martin Plajner, Logio

Churn rate as the rate of customers dropping out of a service, is an important indicator for many modern companies. Its detection thus plays a key part in customer retention and satisfaction. The detection is usually done by an AI/ML model based on neural networks or gradient boosted decision trees. Nevertheless, the sole detection of churn happening is not enough by itself. Knowing a customer is likely to drop out, it is necessary to take an action which makes them stay. Such an action might cover various options such as discounts, coupons, direct calls or email campaigns as so on. It is essential to determine the correct cause to avoid unwanted wasted resources. The second task of analysis and prediction of causes which makes a person leave is even harder than the churn detection itself. In this article we cover the process of identifying churn and with such knowledge we provide a Bayesian network based model which provides insights of reasons for such a churn happening. We demonstrate our method and results on a real world example of a taxi-hailing company which is the segment with large customer fluctuation.

A Modular and Adaptive Bayesian System for Spear Phishing Detection

Jan Brabec, Cisco

The growing sophistication of Business Email Compromise (BEC) and spear phishing attacks poses significant challenges to organizations worldwide. The techniques featured in traditional spam and phishing detection are insufficient due to the tailored nature of modern BEC attacks as they often blend in with the regular benign traffic. Recent advances in machine learning, particularly in Natural Language Understanding (NLU), offer a promising avenue for combating such attacks but in a practical system, due to limitations such as data availability, operational costs, verdict explainability requirements or a need to robustly evolve the system, it is essential to combine multiple approaches together. We present CAPE, a comprehensive and efficient system for BEC detection that has been proven in a production environment for a period of over two years. Rather than being a single model, CAPE is a system that combines independent ML models and algorithms detecting BEC-related behaviors across various email modalities such as text, images, metadata and the email’s communication context. This decomposition makes CAPE’s verdicts naturally explainable. We describe the design principles and constraints behind its architecture, as well as the challenges of model design, evaluation and adapting the system continuously through a Bayesian approach that combines limited data with domain knowledge. Furthermore, we showcase recent developments that integrate generative LLMs into the detection pipeline.

Building an Efficient Geometric Deep Learning Pipeline

Idrees Muhammad, Turing Artificial Intelligence

Geometric Deep Learning (GDL) has emerged as a pivotal paradigm, enabling machines to glean profound insights from structured data such as graphs and meshes. This talk delves into the intricacies of constructing a robust and efficient Geometric Deep Learning Pipeline, a key cornerstone for unlocking the potential of geometric data.

The presentation commences with an exploration of the fundamental principles underpinning Geometric Deep Learning, elucidating its unique capacity to process and interpret non-Euclidean data structures. Attendees will gain a comprehensive understanding of how GDL extends the frontiers of traditional deep learning methodologies, offering unparalleled capabilities in tasks ranging from graph classification to 3D shape recognition.


Poster session:

  • Unlocking the Secrets of Singing: A Neural Network for Transcribing Human Sung Voice (Jeremy Cochoy)
  • Navigating Nature's Labyrinth: Designing a Simulation Framework for implementing and testing Autonomous Drone Flight on Forest Roads, using Reinforcement Learning (Alexander Buchelt)
  • Scalable Approximate NonSymmetric Autoencoder for Collaborative Filtering (Martin Spisak, Radek Bartyzal, Antonín Hoskovec, Ladislav Peška, Miroslav Tůma)
  • Evaluating Machine Learning (AI) Approaches for Classification and Clustering of Marine Mammal Vocalizations in Underwater Acoustic Environments (Attila Aradi, Attila Karoly Varga, Peter Takacs)
  • Pioneering Sound-Based Reinforcement Learning in CNC Operations (Jan Tožička, Michal Linda, Marek Bezůšek, Jan Pecka)
  • Feature extraction and validation from genomic copy number profiles in cancer (Adam Streck, Roland F. Schwarz)
  • Auction Algorithms at Play: Search Engines, trivago, OTA and User Experience (Peter Brejcak)

Advanced RAG: Your Company's Ultimate AI Assistant

John Sinderwing, Entecon

With Retrieval-Augmented Generation (RAG), you can ask GPT models questions about your own data and get back answers with references and quotes from it. In other words, your answer generation is augmented with smart retrieval of provided data. Nvidia predicts that RAG will be one of the hottest AI topics in 2024. Applying the latest LLMs in your business can be tricky. You most likely have a lot of classified or simply niche, company-specific data that you want the LLM to take into consideration. Additionally, you can not have your LLM spout damaging hallucinations. How much background context you can send into LLMs is currently bounded by both technical and cost limitations. This is where RAG systems excel. A RAG system can automatically search your internal data for the most relevant data to answer each query and force the LLM to base its answer entirely on data provided by your defined tools. RAG is a relatively new and fast evolving technique with lots of contributors. Some common pitfalls and issues tend to crop up regardless of industry. That is why we want to share our progress with everyone so that you, too, can create actual business value with LLMs and contribute to the field. In this talk we will cover the most common problems and their solutions. For example: * asking questions about multiple sources at the same time, * adding more tools and subquerying (e.g. internal databases, web search, API calls), * handling different types of data (SQL, JSON etc.), * validating your system output (ROUGE, BERTScore, Gaia etc.) and * prohibiting topics with guardrails (like politics, talking about competitors etc.) The less technically inclined (the technically reclined, if you will), will still enjoy an analogy-heavy presentation filled with pop culture references in order to preserve client anonymity. Neural networks designed for natural language understanding are often inspired by how human brains function. They share many of the same strengths and weaknesses, which makes the whole thing easier to grasp! Entecon is a consulting firm based in Stockholm, Sweden. With a long history in NLP, we have worked with a wide variety of clients and domains. Our AI R&D consultants frequently take experimental ML theory into practice.

Insights into Scam Detection: Training Large Language Models with Limited Datasets

Branislav Bošanský, Gen

The necessity of combating scams resonates strongly in the face of escalating financial losses, reaching billions of dollars worldwide. Scammers use diverse communication channels such as social networks, emails, phone calls, and text messages to contact their victims. This diversity poses a significant technical challenge for designing scam-detection applications. Overcoming these technical obstacles introduces the challenge of determining whether a received message is a scam. While some scams display obvious indicators and are easily detected, categorizing other messages definitively requires additional contextual information, as these messages may not be decidable based solely on the textual content. This presentation provides insights into the methodology employed for training Large Language Models (LLMs) specifically designed for scam detection. Throughout the project, we grappled with the challenge of limited and non-comprehensive datasets, all while meeting accuracy requirements and tight deadlines. We will describe the strategic decisions and considerations taken to navigate through this triad (data scarcity, accuracy demands, and timely delivery), leading to the successful application deployment of robust scam-detection models.

Interpretation of HR data by Large Language Models

Ludek Kopacek, Workday

Large Language Models (LLMs) have emerged as both a technological marvel and a source of fascination in the field of artificial intelligence. While they hold great promise, there is still an incredible journey from toying with LLMs to utilization in enterprise-grade solutions. Our presentation emphasizes the lessons learned and insights gained from practical LLM model productization for a specific HR data interpretation use case. We will share the journey from model selection and fine-tuning to dataset curation and model evaluation. We address issues related to hallucination, testing, model bias, ethical considerations, and data privacy, emphasizing the importance of responsible AI development and deployment.



Murray Campbell, IBM T. J. Watson Research Center
Hamed Valizadegan, NASA
Madalina Ciortan,
Kenji Suzuki, Tokyo Institute of Technology


Have a great time Prague, the city that never sleeps

You can feel centuries of history at every corner in this unique capital. We'll invite you to get a taste of our best pivo (that’s beer in Czech) and then bring you back to the present day to party at one of the local clubs all night long!

Venue ML Prague 2024 will run hybrid, in person and online!

The main conference as well as the workshops will be held at O2 Universum.

We will also livestream the talks for all those participants who prefer to attend the conference online. Our platform will allow interaction with speakers and other participants too. Workshops require intensive interaction and won't be streamed.

Conference building

O2 Universum
Českomoravská 2345/17a, 190 00, Praha 9


O2 Universum
Českomoravská 2345/17a, 190 00, Praha 9

Now or never Registration

Early Bird

Sold Out

  • Conference days € 270
  • Only workshops € 200
  • Conference + workshops € 440


Sold Out

  • Conference days € 290
  • Only workshops € 230
  • Conference + workshops € 490


Sold out

  • Conference days € 320
  • Only workshops € 260
  • Conference + workshops € 520

What You Get

  • Practical and advanced level talks led by top experts.
  • Networking and drinks with speakers and people from all around the world.
  • Delicious food and snacks throughout the conference.

They’re among us We are in The ML Revolution age

Machines can learn. Incredibly fast. Faster than you. They are getting smarter and smarter every single day, changing the world we’re living in, our business and our life. The artificial intelligence revolution is here. Come, learn and make this threat your biggest advantage.

Our Attendees What they say about ML Prague

Thank you to Our Partners

Co-organizing Partner

Venue Partner

Platinum Partners

Gold Partners

Silver Partners

Communities and Further support

Would you like to present your brand to 1000+ Machine Learning enthusiasts? Send us an email at to find out how to become a ML Prague 2024 partner.

Become a partner

Happy to help Contact

If you have any questions about Machine Learning Prague, please e-mail us at


Jiří Materna
Scientific program & Co-Founder

Teresa Pulda
Event production

Gonzalo V. Fernández
Marketing and social media

Jona Azizaj

Ivana Javná
Speaker support

Barbora Toman Hanousková