Accelerating Machine Learning: the hard case of generic models

Accelerating Machine Learning: the hard case of generic models

Following the spreading of applications powered by Machine Learning models, the issues arised on how to engineer their development and deployment. This is particularly problematic when the applications should run under stringent performance requirements and in a complex environment like a cloud infrastructure with diverse hardware resources (CPUs, FPGAs, etc.). We performed initial work on this issue and proposed an approach that achieved a 3x speedup over the common case, suggesting practices that pave the way for developing more systematic guidelines and tools.   By Alberto Scolari PhD Student @ Politecnico di Milano, working on reconfigurable computing systems at NECSTLab   Operational-izing ML: system issues Machine Learning (ML) models are spreading inside companies, as a basis for their business. Nonetheless, applying ML to your business requires theoretical and technical efforts. As willing-to-be system architects,…
Read More
Internet banking fraud analysis and detection

Internet banking fraud analysis and detection

The significant growth of online banking frauds, fueled by the underground economy of malware, raised the need for effective defense systems. As a consequence, in last the years, banks have upgraded their security measures to protect online transactions from frauds.     We propose a novel approach, Banksealer, that models user’s behavior through his or her interaction with the online banking services from different perspectives but with the common goal of recognizing fraudulent activities.   By Michele Carminati Postdoctoral Researcher @ Politecnico di Milano, working on System Security at NECSTLab.   Over the years, Internet banking has grown in popularity. Unfortunately, this has led to an increase of frauds perpetrated through cyber-attacks, resulting in worldwide substantial financial losses. According to Kaspersky Lab, financial malware is evolving through the collaboration between malware creators and…
Read More
Smart wearables: how to stop worrying about the data and love self-tracking

Smart wearables: how to stop worrying about the data and love self-tracking

310 millions wearable devices sold this year, 2 billion people will be using apps to monitor their bodies, by 2019. Everything can be tracked and measured: heart rate, hours of sleep, food calories, exercise, weight... Yet, the majority of users still struggle to make sense of data and abandon them. We propose novel approaches that get back on the Self in “quantified self” and transform self-tracking in actionable empower.   By Luca Cerina Research Assistant @ Politecnico di Milano, working on wearable devices and biomed applications at NECSTLab. The unmet promise In the recent years, the market for self-tracking apps and wearable devices skyrocketed, with industries and media sharing promises of Health and Wellness waiting just a click away from us, preaching the advent of data-sharing products that will solve…
Read More
The CAOS framework: democratize the acceleration of compute intensive applications on FPGA

The CAOS framework: democratize the acceleration of compute intensive applications on FPGA

The increasing demand for computing power in fields as Biology, Machine Learning and Physics is pushing the adoption of reconfigurable hardware as FPGA in order to keep up with the required performance level at a sustainable power consumption. CAOS is a framework to help the application designer in identifying acceleration opportunities and guides through the implementation of the final FPGA-based system.   By Marco Rabozzi  PhD Student @ Politecnico di Milano, working on reconfigurable computing systems at NECSTLab As of today, the progress in many fields of the science is somewhat connected to the amount of available computing power that we have. For instance, higher amount of computing power translates into the capability of simulate a larger amount of neurons and synopsis within a brain, simulate the behavior of more complex physical…
Read More
Do you WannaCry? Protecting from Modern Ransomware Attacks

Do you WannaCry? Protecting from Modern Ransomware Attacks

Infamous ransomware families, malicious programs that encrypt victims’ files preventing legitimate access until a ransom is paid, had a drastic impact in the past years. We proposed a novel approach, ShieldFS, that is able to detect malicious behaviors and revert the effects of ransomware attacks, which means no files lost for the end users!   By Andrea Continella PhD Student @ Politecnico di Milano, soon joining the SecLab @ University of California, Santa Barbara as a Postdoc researcher. In the last year, ransomware has been one the most dangerous Internet threat. Preventive and reactive security measures can only partially mitigate the damage caused by modern ransomware attacks. The remarkable amount of illicit profit and the cybercriminals' increasing interest in ransomware schemes demonstrate that current defense solutions are failing, and a large number…
Read More
“Hackathon on Rare Diseases” – Winner projects

“Hackathon on Rare Diseases” – Winner projects

On 30 September 2017 the projects "Spy:hunt, cyber-physical gaming platform for hospitalized kids" (by Daniele Enoletto, Chiara Di Vece, Filippo Bracco and Riccardo Vailati) and "Preventable, a device and an app service aimed to prevent illness attacks or recidivism"(by Shrine Graai, Luca Cerina, Luca Paccani, Philip Grasselli) won the “Hackathon on rare diseases” organized by Giorgia Zunino at the Health Forum in Florence. The teams were a mix of young NECSTLab engineers and LABA designers (from Brescia) who had been working for two days to find new ideas that could help to improve prevention, quality of life and research for the rare diseases. Winners abstract: SPY:HUNT More than 70% of people affected by rare diseases are children. This condition often forces them to endure long and frequent therapy sessions that…
Read More
“Switch2Product Innovation Challenge” – NECST Winner

“Switch2Product Innovation Challenge” – NECST Winner

DockerCap's project by Rolando Brondolin has won the first award of the “Switch2Product Innovation Challenge” (category: “Idee Imprenditoriali"), an idea by PoliHub, Deloitte Italia and The Technology Transfer office of Politecnico di Milano, that was created to enhance research-driven technologies, develop new entrepreneurship and promote innovative business solutions. Project’s abstract: The exponential growth of the cloud-based solutions offered both by market leaders and startups is leading to a renewed interest on techniques able to manage and optimize cloud-based applications. The Docker container technology, in particular, proposes novel challenges in an extremely open market, from the correct performance management, applications metering and monitoring and power-aware optimizations. DockerCap optimizes power consumption of infrastructures and applications, still guaranteeing the performance the applications and customers requires.
Read More
“XILINX Open Hardware 2017” – PhD category Winner

“XILINX Open Hardware 2017” – PhD category Winner

At NECSTLab, we are happy for the new achievement: on 31 August, at the XILINX Open Hardware 2017, the project "N-BObbiDY boo: Magic Acceleration of N-Body Simulation" by Emanuele Del Sozzo, Marco Rabozzi, Marco Nanni, won the PhD category. An abstract from this project: N-Body simulation is the process that describes the evolution of a system of forces that is composed of N particles. It is a highly important process in physics as it is as general that it can describe the behavior of multiple particles in fields that range from the astrophysics to molecular dynamics. The main problem with this algorithm is that its most accurate approach, called the All-Pairs method, is particularly compute intensive and pure software implementation on CPUs are inefficient both from a performance and power…
Read More