Baris Kurtkaya

Baris Kurtkaya is a PhD student in Electrical and Computer Engineering at UC Santa Barbara, where they study how neural networks use latent-space geometry to support capabilities such as memory and generalization. Their research includes work on the dynamical mechanisms behind short-term memory in recurrent neural networks, published at ICML 2025.

Career

Cosigned byLoading...

CosignedSign in to view

Elsewhere

Koç University

MSc in Computer Science and Engineering

University of California - Santa Barbara

Products & media

bariscankurtkaya.github.io

Dynamical Phases of Short-Term Memory Mechanisms in RNNs

Explore research on how slow-point manifolds and limit cycles can support short-term memory in recurrent neural networks.

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Profile name

Company name

Company name

Baris Kurtkaya

Koç University

University of California - Santa Barbara

Products & media

Dynamical Phases of Short-Term Memory Mechanisms in RNNs

Curated Lists

List Inclusions