no code implementations • 18 Mar 2024 • Armin Karamzade, KyungMin Kim, Montek Kalsi, Roy Fox
In standard Reinforcement Learning settings, agents typically assume immediate feedback about the effects of their actions after taking them.
no code implementations • 22 Feb 2024 • Dmitrii Krylov, Armin Karamzade, Roy Fox
Our method, Moonwalk, has a time complexity linear in the depth of the network, unlike the quadratic time complexity of na\"ive forward, and empirically reduces computation time by several orders of magnitude without allocating more memory.
no code implementations • 2 Dec 2022 • Francesco Malandrino, Giuseppe Di Giacomo, Armin Karamzade, Marco Levorato, Carla Fabiana Chiasserini
To make machine learning (ML) sustainable and apt to run on the diverse devices where relevant data is, it is essential to compress ML models as needed, while still meeting the required learning quality and time performance.
no code implementations • 27 Nov 2020 • Armin Karamzade, Amir Najafi, Seyed Abolfazl Motahari
In this paper, we extend a class of celebrated regularization techniques originally proposed for feed-forward neural networks, namely Input Mixup (Zhang et al., 2017) and Manifold Mixup (Verma et al., 2018), to the realm of Recurrent Neural Networks (RNN).
1 code implementation • 26 Dec 2018 • Mostafa Tavassolipour, Armin Karamzade, Reza Mirzaeifard, Seyed Abolfazl Motahari, Mohammad-Taghi Manzuri Shalmani
In Uncoded method, data symbols are scaled and transmitted through the channel.