no code implementations • 12 Feb 2024 • Elena Ferro, Athanasios Vasilopoulos, Corey Lammie, Manuel Le Gallo, Luca Benini, Irem Boybat, Abu Sebastian
Analog In-Memory Computing (AIMC) is an emerging technology for fast and energy-efficient Deep Learning (DL) inference.
no code implementations • 30 Jan 2024 • Samuele Ruffino, Geethan Karunaratne, Michael Hersche, Luca Benini, Abu Sebastian, Abbas Rahimi
Classification based on Zero-shot Learning (ZSL) is the ability of a model to classify inputs into novel classes on which the model has not previously seen any training examples.
Ranked #3 on Zero-Shot Learning on CUB-200-2011
1 code implementation • 29 Jan 2024 • Michael Hersche, Francesco Di Stefano, Thomas Hofmann, Abu Sebastian, Abbas Rahimi
Abstract reasoning is a cornerstone of human intelligence, and replicating it with artificial intelligence (AI) presents an ongoing challenge.
no code implementations • 9 Dec 2023 • Aleksandar Terzic, Michael Hersche, Geethan Karunaratne, Luca Benini, Abu Sebastian, Abbas Rahimi
We build upon their approach by replacing the linear recurrence with a special temporal convolutional network which permits larger receptive field size with shallower networks, and reduces the computational complexity to $O(L)$.
1 code implementation • NeurIPS 2023 • Nicolas Menet, Michael Hersche, Geethan Karunaratne, Luca Benini, Abu Sebastian, Abbas Rahimi
MIMONets augment various deep neural network architectures with variable binding mechanisms to represent an arbitrary number of inputs in a compositional data structure via fixed-width distributed representations.
1 code implementation • 18 Jul 2023 • Manuel Le Gallo, Corey Lammie, Julian Buechel, Fabio Carta, Omobayode Fagbohungbe, Charles Mackin, Hsinyu Tsai, Vijay Narayanan, Abu Sebastian, Kaoutar El Maghraoui, Malte J. Rasch
In this tutorial, we provide a deep dive into how such adaptations can be achieved and evaluated using the recently released IBM Analog Hardware Acceleration Kit (AIHWKit), freely available at https://github. com/IBM/aihwkit.
1 code implementation • 17 May 2023 • Hadjer Benmeziane, Corey Lammie, Irem Boybat, Malte Rasch, Manuel Le Gallo, Hsinyu Tsai, Ramachandran Muralidhar, Smail Niar, Ouarnoughi Hamza, Vijay Narayanan, Abu Sebastian, Kaoutar El Maghraoui
Digital processors based on typical von Neumann architectures are not conducive to edge AI given the large amounts of required data movement in and out of memory.
no code implementations • 24 Mar 2023 • Michael Hersche, Aleksandar Terzic, Geethan Karunaratne, Jovin Langenegger, Angéline Pouget, Giovanni Cherubini, Luca Benini, Abu Sebastian, Abbas Rahimi
Secondly, the proposed factorizer maintains a high accuracy when queried by noisy product vectors generated using deep convolutional neural networks (CNNs).
no code implementations • 16 Feb 2023 • Malte J. Rasch, Charles Mackin, Manuel Le Gallo, An Chen, Andrea Fasoli, Frederic Odermatt, Ning li, S. R. Nandakumar, Pritish Narayanan, Hsinyu Tsai, Geoffrey W. Burr, Abu Sebastian, Vijay Narayanan
Analog in-memory computing (AIMC) -- a promising approach for energy-efficient acceleration of deep learning workloads -- computes matrix-vector multiplications (MVMs) but only approximately, due to nonidealities that often are non-deterministic or nonlinear.
1 code implementation • 9 Nov 2022 • Jovin Langenegger, Geethan Karunaratne, Michael Hersche, Luca Benini, Abu Sebastian, Abbas Rahimi
Disentanglement of constituent factors of a sensory signal is central to perception and cognition and hence is a critical task for future artificial intelligence systems.
no code implementations • 21 Sep 2022 • Dominique J. Kösters, Bryan A. Kortman, Irem Boybat, Elena Ferro, Sagar Dolas, Roberto de Austri, Johan Kwisthout, Hans Hilgenkamp, Theo Rasing, Heike Riel, Abu Sebastian, Sascha Caron, Johan H. Mentink
The massive use of artificial neural networks (ANNs), increasingly popular in many areas of scientific computing, rapidly increases the energy consumption of modern high-performance computing systems.
no code implementations • 14 Jul 2022 • Geethan Karunaratne, Michael Hersche, Jovin Langenegger, Giovanni Cherubini, Manuel Le Gallo-Bourdeau, Urs Egger, Kevin Brew, Sam Choi, INJO OK, Mary Claire Silvestre, Ning li, Nicole Saulnier, Victor Chan, Ishtiaq Ahsan, Vijay Narayanan, Luca Benini, Abu Sebastian, Abbas Rahimi
We demonstrate for the first time how the EM unit can physically superpose multiple training examples, expand to accommodate unseen classes, and perform similarity search during inference, using operations on an IMC core based on phase-change memory (PCM).
2 code implementations • CVPR 2022 • Michael Hersche, Geethan Karunaratne, Giovanni Cherubini, Luca Benini, Abu Sebastian, Abbas Rahimi
Moreover, it is imperative that such learning must respect certain memory and computational constraints such as (i) training samples are limited to only a few per class, (ii) the computational cost of learning a novel class remains constant, and (iii) the memory footprint of the model grows at most linearly with the number of classes observed.
Ranked #4 on Few-Shot Class-Incremental Learning on mini-Imagenet
continual few-shot learning Few-Shot Class-Incremental Learning +1
no code implementations • 11 Mar 2022 • Denis Kleyko, Geethan Karunaratne, Jan M. Rabaey, Abu Sebastian, Abbas Rahimi
Memory-augmented neural networks enhance a neural network with an external key-value memory whose complexity is typically dominated by the number of support vectors in the key memory.
1 code implementation • 9 Mar 2022 • Michael Hersche, Mustafa Zeqiri, Luca Benini, Abu Sebastian, Abbas Rahimi
Compared to state-of-the-art deep neural network and neuro-symbolic approaches, end-to-end training of NVSA achieves a new record of 87. 7% average accuracy in RAVEN, and 88. 1% in I-RAVEN datasets.
no code implementations • 10 Nov 2021 • Chuteng Zhou, Fernando Garcia Redondo, Julian Büchel, Irem Boybat, Xavier Timoneda Comas, S. R. Nandakumar, Shidhartha Das, Abu Sebastian, Manuel Le Gallo, Paul N. Whatmough
We also describe AON-CiM, a programmable, minimal-area phase-change memory (PCM) analog CiM accelerator, with a novel layer-serial approach to remove the cost of complex interconnects associated with a fully-pipelined design.
1 code implementation • 5 Apr 2021 • Malte J. Rasch, Diego Moreda, Tayfun Gokmen, Manuel Le Gallo, Fabio Carta, Cindy Goldberg, Kaoutar El Maghraoui, Abu Sebastian, Vijay Narayanan
We introduce the IBM Analog Hardware Acceleration Kit, a new and first of a kind open source toolkit to simulate analog crossbar arrays in a convenient fashion from within PyTorch (freely available at https://github. com/IBM/aihwkit).
no code implementations • 5 Oct 2020 • Geethan Karunaratne, Manuel Schmuck, Manuel Le Gallo, Giovanni Cherubini, Luca Benini, Abu Sebastian, Abbas Rahimi
Traditional neural networks require enormous amounts of data to build their complex mappings during a slow training procedure that hinders their abilities for relearning and adapting to new data.
Few-Shot Image Classification Vocal Bursts Intensity Prediction
no code implementations • 15 Sep 2020 • Timoleon Moraitis, Abu Sebastian, Evangelos Eleftheriou
Biological neurons and their in-silico emulations for neuromorphic artificial intelligence (AI) use extraordinarily energy-efficient mechanisms, such as spike-based communication and local synaptic plasticity.
no code implementations • 30 Apr 2020 • Adnan Mehonic, Abu Sebastian, Bipin Rajendran, Osvaldo Simeone, Eleni Vasilaki, Anthony J. Kenyon
Machine learning, particularly in the form of deep learning, has driven most of the recent fundamental developments in artificial intelligence.
no code implementations • 8 Apr 2020 • Ana Stanojevic, Giovanni Cherubini, Timoleon Moraitis, Abu Sebastian
In this paper, we propose a system for file classification in large data sets based on spiking neural networks (SNNs).
no code implementations • 25 Mar 2020 • Vinay Joshi, Geethan Karunaratne, Manuel Le Gallo, Irem Boybat, Christophe Piveteau, Abu Sebastian, Bipin Rajendran, Evangelos Eleftheriou
Strategies to improve the efficiency of MVM computation in hardware have been demonstrated with minimal impact on training accuracy.
1 code implementation • 5 Mar 2020 • Kornilios Kourtis, Martino Dazzi, Nikolas Ioannou, Tobias Grosser, Abu Sebastian, Evangelos Eleftheriou
Computational memory (CM) is a promising approach for accelerating inference on neural networks (NN) by using enhanced memories that, in addition to storing data, allow computations on them.
no code implementations • 8 Jun 2019 • Martino Dazzi, Abu Sebastian, Pier Andrea Francese, Thomas Parnell, Luca Benini, Evangelos Eleftheriou
We show that this communication fabric facilitates the pipelined execution of all state of-the-art CNNs by proving the existence of a homomorphism between one graph representation of these networks and the proposed graph topology.
no code implementations • 7 Jun 2019 • Vinay Joshi, Manuel Le Gallo, Irem Boybat, Simon Haefeli, Christophe Piveteau, Martino Dazzi, Bipin Rajendran, Abu Sebastian, Evangelos Eleftheriou
In-memory computing is a promising non-von Neumann approach where certain computational tasks are performed within memory units by exploiting the physical attributes of memory devices.
Emerging Technologies
no code implementations • 4 Jun 2019 • Geethan Karunaratne, Manuel Le Gallo, Giovanni Cherubini, Luca Benini, Abbas Rahimi, Abu Sebastian
Hyperdimensional computing (HDC) is an emerging computational framework that takes inspiration from attributes of neuronal circuits such as hyperdimensionality, fully distributed holographic representation, and (pseudo)randomness.
no code implementations • 28 May 2019 • S. R. Nandakumar, Irem Boybat, Manuel Le Gallo, Evangelos Eleftheriou, Abu Sebastian, Bipin Rajendran
Combining the computational potential of supervised SNNs with the parallel compute power of computational memory, the work paves the way for next-generation of efficient brain-inspired systems.
no code implementations • 11 Jan 2019 • Bipin Rajendran, Abu Sebastian, Michael Schmuker, Narayan Srinivasa, Evangelos Eleftheriou
In this paper, we review some of the architectural and system level design aspects involved in developing a new class of brain-inspired information processing engines that mimic the time-based information encoding and processing aspects of the brain.
no code implementations • 17 Jun 2017 • Timoleon Moraitis, Abu Sebastian, Irem Boybat, Manuel Le Gallo, Tomas Tuma, Evangelos Eleftheriou
However, some spike-timing-related strengths of SNNs are hindered by the sensitivity of spike-timing-dependent plasticity (STDP) rules to input spike rates, as fine temporal correlations may be obstructed by coarser correlations between firing rates.
no code implementations • 16 Jan 2017 • Manuel Le Gallo, Abu Sebastian, Roland Mathis, Matteo Manica, Heiner Giefers, Tomas Tuma, Costas Bekas, Alessandro Curioni, Evangelos Eleftheriou
As CMOS scaling reaches its technological limits, a radical departure from traditional von Neumann systems, which involve separate processing and memory units, is needed in order to significantly extend the performance of today's computers.
Emerging Technologies