HEATING UP DECISION BOUNDARIES: ISOCAPACITORY SATURATION, ADVERSARIAL SCENARIOS AND GENERALIZATION BOUNDS
May 3, 2021
Speakers
Bogdan Georgiev
Speaker · 0 followers
Lukas Franken
Speaker · 0 followers
Mayukh Mukherjee
Speaker · 0 followers
About
In the present work we study classifiers' decision boundaries via Brownian motion processes in ambient data space and associated probabilistic techniques. Intuitively, our ideas correspond to placing a heat source at the decision boundary and observing how effectively the sample points warm up. We are largely motivated by the search for a soft measure that sheds further light on the decision boundary's geometry. En route, we bridge aspects of potential theory and geometric analysis (Maz'ya 2011, Grigor'Yan and Saloff-Coste 2002) with active fields of ML research such as adversarial examples and generalization bounds. First, we focus on the geometric behavior of decision boundaries in the light of adversarial attack/defense mechanisms. Experimentally, we observe a certain capacitory trend over different adversarial defense strategies: decision boundaries locally become flatter as measured by isoperimetric inequalities (Ford et al 2019); however, our more sensitive heat-diffusion metrics extend this analysis and further reveal that some non-trivial geometry invisible to plain distance-based methods is still preserved. Intuitively, we provide evidence that the decision boundaries nevertheless retain many persistent "wiggly and fuzzy" regions on a finer scale. Second, we show how Brownian hitting probabilities translate to soft generalization bounds which are in turn connected to compression and noise stability (Arora et al 2018), and these bounds are significantly stronger if the decision boundary has controlled geometric features.In the present work we study classifiers' decision boundaries via Brownian motion processes in ambient data space and associated probabilistic techniques. Intuitively, our ideas correspond to placing a heat source at the decision boundary and observing how effectively the sample points warm up. We are largely motivated by the search for a soft measure that sheds further light on the decision boundary's geometry. En route, we bridge aspects of potential theory and geometric analysis (Maz'ya 2011,…
About ICLR 2021
The International Conference on Learning Representations (ICLR) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence called representation learning, but generally referred to as deep learning. ICLR is globally renowned for presenting and publishing cutting-edge research on all aspects of deep learning used in the fields of artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, text understanding, gaming, and robotics.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Panel: Manifold Learning 2.0
Watch later
Oral Session 11 - QA 1
Watch later
Lizhen Nie, …
Biased Client Selection in Federated Learning
Watch later
Gauri Joshi, …
Panel Discussion #1
Watch later
Alina Oprea, …
Reset-Free Lifelong Learning with Skill-Space Planning
Watch later
Kevin Lu, …
