Master thesis project: Formal verification of Machine Learning based models

Rekryterare
Ericsson
Plats
Stockholm, SE
Annonserat
den 5 oktober 2021
Stängs
den 27 oktober 2021
Ref:
595927-en_US
Anställningsform
Tillsvidare
Sysselsättning
Heltid
Why is Ericsson a good place to work?

Ericsson enables communications service providers to capture the full value of connectivity. The company's portfolio spans Networks, Digital Services, Managed Services, and Emerging Business and is designed to help our customers go digital, increase efficiency, find new revenue streams, and create new user experiences. Ericsson's investments in innovation have delivered the benefits of telephony and mobile broadband to billions of people around the world ensuring our solutions - and our customers - are at the forefront of innovation. We support networks that connect more than 2.5 billion subscribers. With over 100,000 employees and customers in 180 countries, we combine global scale with technology and service leadership. 40 percent of the world's mobile traffic is carried over an Ericsson network. And, our Technology for Good and Connect to Learn programs include creating technology that makes it easier to save lives, feed societies, bring technology to emerging markets and connectivity to remote areas, and grow businesses and prosperity.

At Ericsson, we give our employees the freedom to think big and navigate their career, on a global scale. We create technology that helps others, from helping people enjoy their favourite content to helping people recover from natural disasters by enabling better communications between rescue workers. Your ideas and innovations can turn into achievements that impact society and change the world, creating new connections, new possibilities, and new capabilities. We find that Ericsson is at its best when we bring together the diverse skills of our people. Working across business areas, across cultures, across geographical borders, across technical disciplines. More often than not, across ground-breaking solutions. Next generation technology can be staggeringly complex. But the simpler it is to use; the more people benefit from it. Join us and help build technology that makes it simple to connect with information, business, societies, and each other.

Our Exciting Opportunity

Machine learning is making inroads to every industry and a variety of applications. However, the rich and well-studied processes and methods of software engineering like testing and verification to ensure the quality of software products is still in a nascent stage so far as ML models are concerned. Therefore, apart from the metrics such as accuracy, AUC etc. obtained from test datasets, there is no formal guarantee for safety of these ML models. As a consequence, trustworthiness of ML models, especially in safety- and business- critical systems, is low, which also impacts the rapid adoption of ML models.

In this context, formal specification and verification of machine learning models is gaining momentum because of their proven strength in hardware and software verification. Recently, there has been a large number of publications and tools to address the verification of neural networks, and to a much smaller extent, of models such as decision trees, recurring NN's and variants, and deep reinforcement learning models. A comprehensive collection of the tools with pointers to the associated papers can be found in [1][2].

The tasks of this thesis will be to study formal verification for different types of ML models, and implement formal verification tool(s) for these ML models, and specifications and evaluate them for telecom use cases. It is expected that the scope of the tools are limited to certain subclasses of models and specifications. Then, the thesis should implement necessary extensions to address the gaps. We are looking for 1-2 open-minded students who seek a challenging research work with the freedom to propose and develop their own ideas. The thesis project will start in January.

You will

  • Perform selective literature review, identifying relevant concepts and algorithms for formal verification of RNN's and variants, and decision trees.
  • Implement a platform with well-defined interfaces to allow specifying properties in a standard format, converting models into standard format and applying the algorithms
  • Evaluate the tools with ML models for telecom for different properties
  • Identify gaps and implement necessary extensions to the tools.


To be successful in the role you must have

  • MSc studies in Computer Science, AI, Electrical and Computer Engineering or similar.
  • Excellent programming skills in Python or Java or Prolog or C/C++.
  • Knowledge of AI (specifically, ML), and experience with ML model building and manipulation (sklearn, Keras, etc )
  • Logics and logic-based methods e.g. satisfiability (SAT)
  • Linear Programming (LP), MILP, Constraint programming (desirable)
  • Interest in formal verification
  • Passion for AI research and/or building end-to-end AI prototype applications.
  • Fluency in English.


Application

Please send in your application in English as soon as possible. If you have any questions you are welcome to contact recruiter Jacob Elmeljung at: jacob.elmeljung@ericsson.com

What´s in it for you?

Here at Ericsson, our culture is built on over a century of courageous decisions. With us, you will no longer be dreaming of what the future holds - you will be redefining it. You won't develop for the status quo, but will build what replaces it. Joining us is a way to move your career in any direction you want; with hundreds of career opportunities in locations all over the world, in a place where co-creation and collaboration are embedded into the walls. You will find yourself in a speak-up environment where empathy and humanness serve as cornerstones for how we work, and where work-life balance is a priority. Welcome to an inclusive, global company where your opportunity to make an impact is endless.

What happens once you apply?

To prepare yourself for next steps, please explore here: https://www.ericsson.com/en/careers/job-opportunities/hiring-process

[1] https://sites.google.com/view/vnn20/vnncomp , https://github.com/stanleybak/vnncomp2021

[2] https://www.diva-portal.org/smash/get/diva2:1453682/FULLTEXT01.pdf

Do you believe that an organization fostering an environment of cooperation and collaboration to execute with speed creates better business value? Do you value a culture of humanness, where fact based decisions are important and our people are encouraged to speak up? Do you believe that diverse, inclusive teams drive performance and innovation? At Ericsson, we do.

We provide equal employment opportunities without regard to race, color, gender, sexual orientation, transgender status, gender identity and/or expression, marital status, pregnancy, parental status, religion, political opinion, nationality, ethnic background, social origin, social status, indigenous status, disability, age, union membership or employee representation and any other characteristic protected by local law or Ericsson's Code of Business Ethics.

Primary country and city: Sweden (SE) || || Stockholm || [[mfield2]]

Req ID: 595927

Liknande jobb

Liknande jobb