Alberto Speranzon presents On the Stratified Space of Some Deep Learning Models
On 2025-12-09 11:00:00 at E112, Karlovo náměstí 13, Praha 2
Abstract: In this talk, we will present recent work exploring the implicit
representations that deep networks develop in their latent spaces.
Specifically,
we will show how a transformer‑based reinforcement‑learning (RL) agent
tasked with solving simple games embeds visual observations not on a smooth
manifold but on a stratified embedding space whose local dimensionality varies
with sub‑strategic execution and environment complexity. We will argue that
the ambient space of these games is itself stratified, making a stratified
latent representation a natural match. These insights into latent‑space
structure may, in turn, lead to new understandings of how deep models learn and
generalize.
Bio: Alberto Speranzon joined Lockheed Martin’s Advanced
Technology Labs
in March 2022 as Chief Scientist for Autonomy. At Lockheed Martin, his work
focuses on embedding neuro‑symbolic methods into autonomous‑system design,
emphasizing scalability, compositional reasoning, and assurance. Prior to that,
he served as a Technical Fellow at Honeywell Aerospace, leading research on
perception, navigation, world‑modeling, and decision‑making for
next‑generation aerospace systems. He has held editorial positions with IEEE
Transactions on Control Systems Technology and currently serves as an Associate
Editor for the IEEE Open Journal of Control Systems.
representations that deep networks develop in their latent spaces.
Specifically,
we will show how a transformer‑based reinforcement‑learning (RL) agent
tasked with solving simple games embeds visual observations not on a smooth
manifold but on a stratified embedding space whose local dimensionality varies
with sub‑strategic execution and environment complexity. We will argue that
the ambient space of these games is itself stratified, making a stratified
latent representation a natural match. These insights into latent‑space
structure may, in turn, lead to new understandings of how deep models learn and
generalize.
Bio: Alberto Speranzon joined Lockheed Martin’s Advanced
Technology Labs
in March 2022 as Chief Scientist for Autonomy. At Lockheed Martin, his work
focuses on embedding neuro‑symbolic methods into autonomous‑system design,
emphasizing scalability, compositional reasoning, and assurance. Prior to that,
he served as a Technical Fellow at Honeywell Aerospace, leading research on
perception, navigation, world‑modeling, and decision‑making for
next‑generation aerospace systems. He has held editorial positions with IEEE
Transactions on Control Systems Technology and currently serves as an Associate
Editor for the IEEE Open Journal of Control Systems.