At a glance
- Opportunity
- Master's thesis
- Location
- Stockholm, SE
- Funding / pay
- Not specified by the organization
- Eligibility
- Check the qualifications in the official posting
Project description
## Join our Team About this opportunity Ericsson is looking for a master’s thesis student in Computer Science or Electrical Engineering with strong deep-learning skills and an interest in sequence modelling, information security, or applied research. Random-number generators are critical components of secure communication and cryptographic systems. NIST SP 800-22 provides a suite of statistical tests for assessing binary sequences, while NIST SP 800-90B defines methods for estimating min-entropy, including Markov and predictor-based estimators. These methods were developed before the emergence of modern machine-learning techniques. This thesis investigates whether compact Transformer-based sequence models, trained from scratch on binary data, can extend and complement established NIST randomness and entropy-evaluation methods. By learning to predict the next bit or block in a sequence, the model performs a generalised prediction and compression task related to methods already used in existing standards. Its prediction performance can therefore be studied as a potential statistical test for detecting dependencies and entropy deficiencies. What you will do: In this thesis, you will conduct a literature review of machine-learning-based extensions and alternatives to established randomness tests. You will design, implement, and compare compact Transformer architectures trained from scratch for next-bit and next-block prediction on binary sequences. You will evaluate the resulting model-based test using weak pseudorandom number generators, NIST SP 800-90A Deterministic Random Bit Generators—including Hash_DRBG, HMAC_DRBG, and CTR_DRBG - and data from a Quantum Random Number Generator available at Ericsson. You will compare the models’ results with established statistical randomness tests and analyse the relationship between prediction performance, compression-based measures, and min-entropy estimates from NIST SP 800-90B, including predictor-based methods such as Lag, MultiMMC, and LZ78Y. The main objective is to determine whether machine-learning-based prediction can reveal statistical dependencies or entropy deficiencies that conventional methods may not detect. Any differences observed between PRNG and QRNG data must be interpreted carefully, as they may arise from device bias, preprocessing, formatting, or other dataset-specific artifacts rather than from a uniquely quantum signature. The skills you bring: • Enrolled in the final year of an MSc programme in Physics, Engineering Physics, Computer Science, Electrical Engineering, Applied Mathematics, Cybersecurity, or a closely related field. • Strong Python programming skills and hands-on experience with PyTorch or TensorFlow. • Ability to design, implement, train, and evaluate neural-network architectures from scratch. • Good understanding of probability, statistics, information theory, and fundamental machine-learning concepts. • Familiarity with data preparation, training and validation workflows, performance evaluation, and reproducible experimentation. • Understanding of binary data representation and sequential data processing. • Experience with Transformers, sequence modelling, natural language processing, or time-series analysis is a strong advantage. • Familiarity with NumPy, pandas, and scikit-learn, or a demonstrated ability and willingness to learn new tools quickly. • Basic knowledge of randomness, entropy estimation, statistical testing, cryptography, or pseudorandom and quantum random-number generation is advantageous. • Proficiency in spoken and written English. • Ability to work independently, manage a 20-week research project, document results, and communicate findings clearly through technical reports and presentations. Why join Ericsson? At Ericsson, you will have an outstanding opportunity. The chance to use your skills and imagination to push the boundaries of what is possible. To build solutions never seen before to some of the world’s toughest problems. You will be challenged, but you won’t be alone. You´ll be joining a team of diverse innovators, all driven to go beyond the status quo to craft what comes next. Encouraging a diverse and inclusive organization is core to our values at Ericsson, that' is why we champion it in everything we do. We truly believe that by collaborating with people with different experiences we drive innovation, which is essential for our future growth. We encourage people from all backgrounds to apply and realize their full potential as part of our Ericsson team. Ericsson is proud to be an Equal Opportunity Employer.
Funding and compensation
Compensation not specified. See the original posting for amounts, duration and conditions.
Eligibility and application requirements
Review the qualifications, research interests and required documents in the description above. The employer or university's original posting is the source for complete eligibility requirements.
Before you apply
0 of 4 readyA preparation checklist, saved on this device. The official posting determines eligibility and required documents.
How to apply
Submit your application through Ericsson's official application process before 20 Mar 2027.
Source and listing information
This opportunity is published by Ericsson. Read the original official posting. Conditions and availability may change; confirm them with the organization. Availability verification refers to the source check, not organization endorsement.
Report an outdated or incorrect listing