Mahum Naseer

Mahum Naseer
Further Staff Projektass. (DK-RES)
Phone +43 1 58801 - 18267
Fax +43 (1) 58801 - 18297

Vienna University of Technology
Institute of Computer Engineering
Embedded Computing Systems
Treitlstrasse 3, 2nd floor, 1040 Wien, Austria

About myself

I am currently a University Research Assistant at Computer Architecture and Robust, Energy-Efficient Technologies (CARE-Tech.), and a PHD student at Technische Universität Wien (TU-Wien), Vienna, Austria. I completed my Masters in Electrical Engineering at National University of Sciences and Technology (NUST), Islamabad, Pakistan, in 2018.


Research and Projects

Over past few decades, machine learning has achieved impressive results in object detection and classification, speech recognition, bio-informatics, and social network filtering. Due to its extraordinary potential, it holds a promising future for many applications like health care and autonomous driving. However, it is easily fooled in the presence of minute perturbations in input. This makes ensuring the robustness of the machine learning systems a critical safety and security concern. The current machine learning verification approaches are either limited by their scalability, or imprecise/incomplete results. My current research mainly lies in formal verification of realistically sized machine learning systems.

My prior projects include:

  • Model checking based approach, FANNet, for the formal analysis of trained neural network, to determine network’s noise tolerance, training bias and input node sensitivity (

  • A framework for formal verification of Error Correction Codes (ECCs) used in memories, using ACL2 theorem prover - 2018

  • Video Skeletonization based on Distance Transform using MatLab - 2017

  • Multi-bit Multiplier and Accumulator circuit design using Electric VLSI Design System - 2016

  • Low cost and power efficient HYbrid Terrestrial and Aerial Quadrotor (HYTAQ), with obstacle detection and surveillance system - 2015

  • Energy Conservation System (prototype for classrooms) based on PIC micro-controller and IR-Sensor - 2013