Semantic Data & ML DevOps; A Cybersecurity Perspective on Adversarial ML

Location in Betahaus: floor 1 Innospace

Talk 1: Semantic Data and ML DevOps : An Industrial IoT case study

Speaker: Semih Korkmaz (Datagraph)

Abstract: ML model development and application faces challenges addressing understandability, debugging, configuration management, and operational support. This talk explores the data strategies that incorporate semantic data, data provenance, schema mutability and the relationship to effective ML development, deployment and operational practices by providing a reference implementation in an industrial IoT application.

Bio: Semih has almost a decade of experience designing and developing machine learning solutions including recommendation engines for Vodafone Mobile TV systems and Industry 4.0 applications such as reinforcement learning models for assembly tasks at Arcelik/Beko. His edge analytics projects on Siemens Systems have become accepted as a ML model deployment strategy at several manufacturing locations. Formerly, Semih was a lecturer at German Research Center for Artificial Intelligence, and a researcher Max Planck Institute.

--

Talk 2: Adversarial Machine Learning: A Cybersecurity Perspective

Speaker: Amit Kushwaha

Abstract: Security and Privacy issues need no introduction. But how exactly is this affecting the field of Machine Learning? This is what this talk will cover. We first expose the attack surface of systems deploying machine learning. We then describe how an attacker may force models to make wrong predictions with very little information about the victim. One such attack can be biometric recognition where fake biometric traits may be exploited to impersonate a legitimate user. We demonstrate that these attacks are practical against existing machine learning as a service platform. Towards the end, we will discuss current research to defend models from such attacks.

Bio: Amit Kushwaha is a Python Backend Engineer in the Pricing and Forecasting Team of Zalando. He works on large-scale Optimal Discount Recommendation. He worked earlier as an ML Engineer in Zomato. His interests are in Deep Learning, Recommendation Systems, NLP and Data Engineering. He works with the Tensorflow, Keras, Pyspark, Airflow, Luigi and Pandas. He dreams to pursue AI as an independent researcher in future.