Call for PhD & Postdoctoral Applications — Academic Year 2026/2027
BBQ Institute invites applications for fully funded Doctoral and Postdoctoral Fellowships in Machine Learning, Quantum Computing, and Distributed Systems. Applications close on September 30, 2026.
Project VERIPROVE: Automated Verification of Distributed Neural Computation
A major research initiative developing sound, scalable formal verification tools and cryptographic proofs for deep neural network execution across distributed computing nodes.
Principal Investigator
Prof. dr hab. inż. Janusz Kowalczyk
Laboratory
Laboratory for Machine Cognition & Neural Systems (MCNC Lab)
Funding Agency
National Science Centre (NCN) — OPUS 24 Scheme
Grant ID
2024/53/B/ST6/01928
Allocated Budget
PLN 2,840,000 (€640,000)
Project Period
2024–2027
Scientific Objective & Core Research Questions
How can deep neural networks deployed across untrusted distributed environments be verified for correctness, robustness, and mathematical safety with minimal computational overhead?
Work Packages & Methodological Roadmap
Constructing polynomial certificates for neural activation bounds using semi-definite programming and branch-and-bound optimization.
Designing zero-knowledge verification protocols (zk-SNARKs) tailored for transformer matrix multiplications and attention maps.
Deploying automated verification suites across 64 cluster nodes in the CeNT datacenter to evaluate throughput and proof generation latency.
Project Deliverables & Software Artefacts
VERIPROVE-Core: Open-source neural verification engine written in Rust and C++ with Python bindings.
Benchmark Dataset of 100,000 certified invariant problems for deep vision and language models.
Containerized reproducible test suite hosted on the BBQ Institute open science repository.
Project Description & Objectives
The widespread adoption of foundation models in mission-critical applications—from autonomous robotics to clinical diagnosis—demands mathematical guarantees rather than purely empirical testing. Project VERIPROVE tackles this challenge by creating a unified theoretical and algorithmic foundation for verifying complex neural models.
Work Packages
WP1: Scalable Interval & Polyhedral Abstraction
Developing tight numeric abstractions that propagate input bounds through non-linear activation functions without exponential state explosion.
WP2: Cryptographic Execution Provenance
Engineering lightweight cryptographic attestation proofs that prove a cloud computing instance executed a specific neural network without altering weights or precision.
WP3: Integration with Interactive Theorem Provers
Connecting neural verification routines directly to Lean 4 and Coq proof assistants, enabling end-to-end formal mathematical certification.
Project Milestones & Reporting
Annual scientific progress report submitted to NCN in January 2026.
International workshop planned for Autumn 2026 in Warsaw.
Mark D. Wilkinson, Michel Dumontier, Tomasz Wiśniewski, Mateusz Wójcik, Barend Mons
Scientific Data (Nature Springer) Vol. 11(1), pp. 18-34(2024). DOI: 10.1038/sdata.2016.18
This foundational work establishes actionable principles ensuring that digital research objects—including datasets, algorithms, and computational workflows—are Findable, Accessible, Interoperable, and Reusable (FAIR) for both humans and automated computational agents.
An extensive review of scientific methodologies, proposing concrete institutional measures to improve transparency, reproducibility, and computational integrity across experimental and data-driven sciences.