Anirban Chowdhury

Yasir Latif, Anirban Chowdhury, Samya Bagchi

Trustless tracking of Resident Space Objects (RSOs) is crucial for Space Situational Awareness (SSA), especially during adverse situations. The importance of transparent SSA cannot be overstated, as it is vital for ensuring space safety and security. In an era where RSO location information can be easily manipulated, the risk of RSOs being used as weapons is a growing concern. The Tracking Data Message (TDM) is a standardized format for broadcasting RSO observations. However, the varying quality of observations from diverse sensors poses challenges to SSA reliability. While many countries operate space assets, relatively few have SSA capabilities, making it crucial to ensure the accuracy and reliability of the data. Current practices assume complete trust in the transmitting party, leaving SSA capabilities vulnerable to adversarial actions such as spoofing TDMs. This work introduces a trustless mechanism for TDM validation and verification using deep learning over blockchain. By leveraging the trustless nature of blockchain, our approach eliminates the need for a central authority, establishing consensus-based truth. We propose a state-of-the-art, transformer-based orbit propagator that outperforms traditional methods like SGP4, enabling cross-validation of multiple observations for a single RSO. This deep learning-based transformer model can be distributed over a blockchain, allowing interested parties to host a node that contains a part of the distributed deep learning model. Our system comprises decentralised observers and validators within a Proof of Stake (PoS) blockchain. Observers contribute TDM data along with a stake to ensure honesty, while validators run the propagation and validation algorithms. The system rewards observers for contributing verified TDMs and penalizes those submitting unverifiable data.

David Layden, Ryan Sweke, Vojtěch Havlíček et al.

Flow models are a cornerstone of modern machine learning. They are generative models that progressively transform probability distributions according to learned dynamics. Specifically, they learn a continuous-time Markov process that efficiently maps samples from a simple source distribution into samples from a complex target distribution. We show that these models are naturally related to the Schrödinger equation, for an unusual Hamiltonian on continuous variables. Moreover, we prove that the dynamics generated by this Hamiltonian can be efficiently simulated on a quantum computer. Together, these results give a quantum algorithm for preparing coherent encodings (a.k.a., qsamples) for a vast family of probability distributions--namely, those expressible by flow models--by reducing the task to an existing classical learning problem, plus Hamiltonian simulation. For statistical problems defined by flow models, such as mean estimation and property testing, this enables the use of quantum algorithms tailored to qsamples, which may offer advantages over classical algorithms based only on samples from a flow model. More broadly, these results reveal a close connection between state-of-the-art machine learning models, such as flow matching and diffusion models, and one of the main expected capabilities of quantum computers: simulating quantum dynamics.

Anirban Chowdhury, Javier Andreu-Perez

In the field of brain-computer interface (BCI) research, the availability of high-quality open-access datasets is essential to benchmark the performance of emerging algorithms. The existing open-access datasets from past competitions mostly deal with healthy individuals' data, while the major application area of BCI is in the clinical domain. Thus the newly proposed algorithms to enhance the performance of BCI technology are very often tested against the healthy subjects' datasets only, which doesn't guarantee their success on patients' datasets which are more challenging due to the presence of more nonstationarity and altered neurodynamics. In order to partially mitigate this scarcity, Clinical BCI Challenge aimed to provide an open-access rich dataset of stroke patients recorded similar to a neurorehabilitation paradigm. Another key feature of this challenge is that unlike many competitions in the past, it was designed for algorithms in both with-in subject and cross-subject categories as a major thrust area of current BCI technology is to realize calibration-free BCI designs. In this paper, we have discussed the winning algorithms and their performances across both competition categories which may help develop advanced algorithms for reliable BCIs for real-world practical applications.