Publications
Bahattin Can Maral, Mehmet Tan · 2022 · IFIP International Conference on Artificial Intelligence Applications and Innovations · Springer International Publishing · pp. 414-425 · Cited by 1
Among the recent developments in bioinformatics and chemogenomics, various deep learning methods have been the most prevalent. This resulted in an over-saturation of powerful models that easily pushed the limits of existing datasets. Subsequently, many novel advancements have been done with improvements to the datasets. Amidst these advancements, researchers of Deep Compound Profiler (DeepCOP) set themselves apart with a novel method of introducing new features whilst keeping the deep learning model relatively basic. In this study, we propose to take this novel method one step further by applying transfer learning between cell lines. In order to better evaluate the benefits of transfer learning, we've introduced 2 drug-based data splits. The transfer learning method, as its core, utilizes the learned knowledge of 'source' cell lines to give a head start to 'target' cell lines.
Bahattin Can Maral · 2022 · arXiv preprint arXiv:2202.11763 · Cited by 18
Super-resolution (SR), the process of obtaining high-resolution images from one or more low-resolution observations of the same scene, has been a very popular topic of research in the last few decades in both signal processing and image processing areas. Due to the recent developments in Convolutional Neural Networks, the popularity of SR algorithms has skyrocketed as the barrier of entry has been lowered significantly. Recently, this popularity has spread into video processing areas to the lengths of developing SR models that work in real-time. In this paper, we compare different SR models that specialize in single image processing and will take a glance at how they evolved to take on many different objectives and shapes over the years.
Rıza Işık, Işıksu Ekşioğlu, Bahattin Can Maral, Benan Bardak, Mehmet Tan · 2020 · 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE) · IEEE · pp. 111-114 · Cited by 1
Understanding the mechanism of action for drugs is vital for drug discovery. Identifying the effect of drugs on gene expression can shed light on the system-side influence of the chemical compounds in biological organisms. In this paper, we propose to use multi-task neural networks to predict chemical induced differential gene expression on cancer cell lines based solely on features of chemicals. Our model predicts differential gene expression identified by a method called Characteristic Direction on a large scale chemical induced gene expression database (LINCS L1000). The results show that the multi-task networks outperform the other single task baselines. We also compare different representations of chemicals and report effect of clustering genes on the prediction performance.
Kimyasalların Gen Düzenleyici Etkilerinin Tahmini için Transfer Öğrenimi
Bahattin Can Maral · 2022 · TOBB ETÜ
Kemogenomik, ilaç tasarımına ve taramaya yardımcı olmak amacıyla biyolojik hedeflerin kimyasal bileşiklere genomik ve/veya proteomik reaksiyonunun incelenmesidir. Kemogenomikteki birçok zorluktan biri, gerçek yaşam deney verilerine bağımlılıktan kaynaklanmaktadır; farklı kimyasal bileşiklerin ve ilaç hedeflerinin kombinasyonu, gerçekçi olmayan sayıda olası deney yaratır ve bu da belirli kimyasallara ve hedeflere yönelik önyargılı veri kümeleriyle sonuçlanmaktadır. Yapay öğrenmedeki son gelişmeler, bu veri kümelerinin sınırlarını kolayca zorlayan güçlü modellerin aşırı doygunluğuyla sonuçlanmıştır. Bu yatkınlıkların etkilerini nötrlemek için, benzer problemlerden bilgi edinme yöntemi olan transfer öğrenmeyi kullanmaktayız. Kemogenomik veri setlerindeki en önemli yanlılık, ilaç hedeflerine yönelik olandır. Bazı hücre dizilerinin erişebilirliği ve önemi, bu deneyler için bir ilaç hedefi olarak kullanılma şansını büyük ölçüde artırırken, diğerlerinin yapay öğrenme modellerini eğitmek için ancak yeterli verisi vardır.
Frequently asked questions
Who is Bahattin Maral?
Bahattin Maral is a Machine Learning Research Engineer based in Barcelona, Spain, with more than 7 years of professional experience across game security, live-service game analytics, data engineering, and applied ML research. He works on Activision’s RICOCHET Anti-Cheat machine learning team, building behavioral and vision-based cheat detection for Call of Duty titles.
What does Bahattin Maral specialise in?
Applied machine learning that has to survive production: behavioral modeling, anomaly detection, and computer vision for game security, plus the large-scale data engineering underneath them. He works end to end — research, feature design, deployment, and monitoring — rather than handing models off at the prototype stage.
How many years of experience does Bahattin Maral have?
More than 7 years of professional experience. He started in 2019 as a data scientist at CBML Labs, then was a senior data scientist at SNG Studios, a data engineer at Digital Legends (an Activision studio), and is now a machine learning research engineer at Activision.
Where does Bahattin Maral work?
At Activision in Barcelona, Spain, on the RICOCHET Anti-Cheat machine learning team. He joined Activision formally in July 2026 after being embedded with the RICOCHET ML team on loan from Digital Legends since May 2025.
What is his experience with anti-cheat and game security machine learning?
He develops behavioral machine learning detections on Activision’s RICOCHET Anti-Cheat team that identify cheating and disruptive play in Call of Duty: Warzone, Black Ops 6, and Black Ops 7, and is extending coverage to Modern Warfare 4 ahead of its October 2026 launch. He also builds computer-vision models that detect cheating from gameplay video, models player behavior at scale to surface automated play and illegitimate account activity, and has contributed detections that powered multiple enforcement ban waves. He leads 3 machine learning projects end to end.
What is his data engineering experience?
At Digital Legends, an Activision studio, he architected the studio’s golden-source data lake and scalable ETL pipelines on Databricks, shipped real-time onboarding pipelines for Call of Duty: Warzone Mobile, built an AI-driven anomaly detection system that auto-files Jira tickets and cut incident reporting time by 90%, optimised Spark queries to reduce dashboard latency for active-user analytics, and delivered self-serve real-time Tableau dashboards.
What measurable results has Bahattin Maral delivered?
A 15% revenue uplift across 40+ mobile games through BigQuery ML pricing models, tripled revenue per ad through mediation optimisation, a 90% reduction in incident reporting time via AI-driven anomaly detection, a 30% reduction in experiment analysis time through a standardised A/B testing framework, 5+ hours per team per week recovered through automated KPI reporting, and anti-cheat detections protecting multiple AAA Call of Duty titles.
Which programming languages and tools does Bahattin Maral use?
Python, SQL, C#, R, Scala, C++, and Java. His tooling includes PyTorch, scikit-learn, Databricks, Apache Spark, BigQuery and BigQuery ML, Google Cloud Platform, Tableau, Unity, and RLCard.
Has Bahattin Maral published research?
Yes. He co-authored "Transfer Learning for Predicting Gene Regulatory Effects of Chemicals" (AIAI 2022, Springer) and "Chemical Induced Differential Gene Expression Prediction on LINCS Database" (IEEE BIBE 2020), and authored the survey "Single Image Super-Resolution Methods: A Survey" (arXiv:2202.11763). His MSc thesis covered transfer learning for chemogenomics.
Does Bahattin Maral have leadership or mentoring experience?
Yes. He leads 3 machine learning projects end to end at Activision, mentored a junior data scientist at SNG Studios, supervised 3 undergraduate researchers at CBML Labs, and was a teaching assistant for Machine Learning, Databases, Object-Oriented Programming, and Discrete Mathematics at TOBB ETÜ.
How do I contact Bahattin Maral?
He is happy to talk about anti-cheat, behavioral modeling, and ML systems for games. Email [email protected] or message him on LinkedIn at linkedin.com/in/bahattinmaral. His resume PDF is at https://punoqun.com/Bahattin-Can-Maral-Resume.pdf and a plain-text profile for language models is at https://punoqun.com/llms.txt.
This is the plain-HTML version of the profile, published for readers, screen readers, and crawlers that do not run JavaScript. Last updated: 2026-08-27.