Search

This work focuses on the challenges of non-IID data and stragglers/dropouts in federated learning. We introduce and explore a privacy-flexible paradigm that models parts of the clients' local data as non-private, offering a more versatile and business-oriented perspective on privacy. Within this framework, we propose a data-driven strategy for mitigating the effects of label heterogeneity and client straggling on federated learning. Our solution combines both offline data sharing and approximate gradient coding techniques. Through numerical simulations using the MNIST dataset, we demonstrate that our approach enables achieving a deliberate trade-off between privacy and utility, leading to improved model convergence and accuracy while using an adaptable portion of non-private data.

Memory disorders are a central factor in the decline of functioning and daily activities in elderly individuals. The confirmation of the illness, initiation of medication to slow its progression, and the commencement of occupational therapy aimed at maintaining and rehabilitating cognitive abilities require a medical diagnosis. The early identification of symptoms of memory disorders, especially the decline in cognitive abilities, plays a significant role in ensuring the well-being of populations. Features related to speech production are known to connect with the speaker's cognitive ability and changes. The lack of standardized speech tests in clinical settings has led to a growing emphasis on developing automatic machine learning techniques for analyzing naturally spoken language. Non-lexical but acoustic properties of spoken language have proven useful when fast, cost-effective, and scalable solutions are needed for the rapid diagnosis of a disease. The work presents an approach related to feature selection, allowing for the automatic selection of the essential features required for diagnosis from the Geneva minimalistic acoustic parameter set and relative speech pauses, intended for automatic paralinguistic and clinical speech analysis. These features are refined into word histogram features, in which machine learning classifiers are trained to classify control subjects and dementia patients from the Dementia Bank's Pitt audio database. The results show that achieving a 75% average classification accuracy with only twenty-five features with the separate ADReSS 2020 competition test data and the Leave-One-Subject-Out cross-validation of the entire competition data is possible. The results rank at the top compared to international research, where the same dataset and only acoustic features have been used to diagnose patients.

Raman spectroscopy is an important tool in the study of vibrational properties and composition of molecules, peptides and even proteins. Raman spectra can be simulated based on the change of the electronic polarizability with vibrations, which can nowadays be efficiently obtained via machine learning models trained on first-principles data. However, the transferability of the models trained on small molecules to larger structures is unclear and direct training on large structures in prohibitively expensive. In this work, we first train two machine learning models to predict polarizabilities of all 20 amino acids. Both models are carefully benchmarked and compared to DFT calculations, with neural network method found to offer better transferability. By combining machine learning models with classical force field molecular dynamics, Raman spectra of all amino acids are also obtained and investigated, showing good agreement with experiments. The models are further extended to small peptides. We find that adding structures containing peptide bonds to the training set greatly improves predictions even for peptides not included in training sets.

We investigate the criminal market dynamics of infostealer malware and publish three evidence datasets on malware infections and trade. We justify the value chain between illicit enterprises using the datasets, compare the prices and added value, and use the value chain to identify the most effective countermeasures. We begin by examining infostealer malware victim logs shared by actors on hacking forums, and extract victim information and mask sensitive data to protect privacy. We find access to these same victims for sale at Genesis Market. This technically sophisticated marketplace provides its own browser to access victim's online accounts. We collect a second dataset and discover that 91% of prices fall between 1--20 US dollars, with a median of 5 US dollars. Database Market sells access to compromised online accounts. We produce yet another dataset, finding 91% of prices fall between 1--30 US dollars, with a median of 7 US dollars., Comment: In The 18th International Conference on Availability, Reliability and Security (ARES 2023), August 29 -- September 1, 2023, Benevento, Italy. 12 pages

Measuring the state of qubits is one of the fundamental operations of a quantum computer. Currently, state-of-the-art high-fidelity single-shot readout of superconducting qubits relies on parametric amplifiers at the millikelvin stage. However, parametric amplifiers are challenging to scale beyond hundreds of qubits owing to practical size and power limitations. Nanobolometers have a multitude of properties that are advantageous for scalability and have recently shown sensitivity and speed promising for qubit readout, but such thermal detectors have not been demonstrated for this purpose. In this work, we utilize an ultrasensitive bolometer in place of a parametric amplifier to experimentally demonstrate single-shot qubit readout. With a readout duration of $13.9~\mu\mathrm{s}$, we achieve a single-shot fidelity of 0.618 which is mainly limited by the energy relaxation time of the qubit, $T_1 = 28~\mu\mathrm{s}$. Without the $T_1$ errors, we find the fidelity to be 0.927. In the future, high-fidelity single-shot readout may be achieved by straightforward improvements to the chip design and experimental setup, and perhaps most interestingly by the change of the bolometer absorber material to reduce the readout time to the hundred-nanosecond level and beyond., Comment: 14 pages, 6 figures

We consider energy stable summation by parts finite difference methods (SBP-FD) for the homogeneous and piecewise homogeneous dynamic beam equation (DBE). Previously the constant coefficient problem has been solved with SBP-FD together with penalty terms (SBP-SAT) to impose boundary conditions. In this work we revisit this problem and compare SBP-SAT to the projection method (SBP-P). We also consider the DBE with discontinuous coefficients and present novel SBP-SAT, SBP-P and hybrid SBP-SAT-P discretizations for imposing interface conditions. Numerical experiments show that all methods considered are similar in terms of accuracy, but that SBP-P can be more computationally efficient (less restrictive time step requirement for explicit time integration methods) for both the constant and piecewise constant coefficient problems., Comment: 28 pages, 5 figures. Submitted to Journal of Computational Physics 26 July 2021

Ultra-reliable low latency communications (URLLC) arose to serve industrial IoT (IIoT) use cases within the 5G. Currently, it has inherent limitations to support future services. Based on state-of-the-art research and practical deployment experience, in this article, we introduce and advocate for three variants: broadband, scalable and extreme URLLC. We discuss use cases and key performance indicators and identify technology enablers for the new service modes. We bring practical considerations from the IIoT testbed and provide an outlook toward some new research directions., Comment: The manuscript is undergoing extensive review

The design, integration, testing, and launch of the first Finnish satellite Aalto-1 is briefly presented in this paper. Aalto-1, a three-unit CubeSat, launched into Sun-synchronous polar orbit at an altitude of approximately 500 km, is operational since June 2017. It carries three experimental payloads: Aalto Spectral Imager (AaSI), Radiation Monitor (RADMON), and Electrostatic Plasma Brake (EPB). AaSI is a hyperspectral imager in visible and near-infrared (NIR) wavelength bands, RADMON is an energetic particle detector and EPB is a de-orbiting technology demonstration payload. The platform was designed to accommodate multiple payloads while ensuring sufficient data, power, radio, mechanical and electrical interfaces. The design strategy of platform and payload subsystems consists of in-house development and commercial subsystems. The CubeSat Assembly, Integration & Test (AIT) followed Flatsat -- Engineering-Qualification Model (EQM) -- Flight Model (FM) model philosophy for qualification and acceptance. The paper briefly describes the design approach of platform and payload subsystems, their integration and test campaigns, and spacecraft launch. The paper also describes the ground segment & services that were developed by the Aalto-1 team., Comment: 41 pages, 16 figures, accepted to Acta Astronautica, 2021

The in-orbit results and lessons learned of the first Finnish satellite Aalto-1 are briefly presented in this paper. Aalto-1, a three-unit CubeSat which was launched in June 2017, performed AaSI (Aalto Spectral Imager), Radiation Monitor (RADMON), and Electrostatic Plasma Brake (EPB) missions. The satellite partly fulfilled its mission objectives and allowed to either perform or attempt the experiments. Although attitude control was partially functional, AaSI and RADMON were able to acquire valuable measurements. EPB was successfully commissioned but the tether deployment was not successful., Comment: 35 pages

The massive amount of data available in operational mobile networks offers an invaluable opportunity for operators to detect and analyze possible anomalies and predict network performance. In particular, application of advanced machine learning (ML) techniques on data aggregated from multiple sources can lead to important insights, not only for the detection of anomalous behavior but also for performance forecasting, thereby complementing classic network operation and maintenance solutions with intelligent monitoring tools. In this paper, we propose a novel framework that aggregates diverse data sets (e.g. configuration, performance, inventory, locations, user speeds) from an operational LTE network and applies ML algorithms to diagnose network issues and analyze their impact on key performance indicators. To this end, pattern identification and time-series forecasting algorithms are used on the ingested data. Results show that proposed framework can indeed be leveraged to automate the identification of anomalous behaviors associated with the spatial-temporal characteristics, and predict customer impact in an accurate manner., Comment: Accepted in IEEE Globecomm 2020 workshop on AI-Enabled 5G/6G Networks: Automation, Openness, and Radio Access