Search Results for author: Evgeny Osipov

Found 11 papers, 3 papers with code

A Survey on Hyperdimensional Computing aka Vector Symbolic Architectures, Part II: Applications, Cognitive Models, and Challenges

no code implementations12 Nov 2021 Denis Kleyko, Dmitri A. Rachkovskij, Evgeny Osipov, Abbas Rahimi

This is Part II of the two-part comprehensive survey devoted to a computing framework most commonly known under the names Hyperdimensional Computing and Vector Symbolic Architectures (HDC/VSA).

A Survey on Hyperdimensional Computing aka Vector Symbolic Architectures, Part I: Models and Data Transformations

no code implementations11 Nov 2021 Denis Kleyko, Dmitri A. Rachkovskij, Evgeny Osipov, Abbas Rahimi

Both names refer to a family of computational models that use high-dimensional distributed representations and rely on the algebraic properties of their key operations to incorporate the advantages of structured symbolic representations and vector distributed representations.

Electrical Engineering

Hyperseed: Unsupervised Learning with Vector Symbolic Architectures

1 code implementation15 Oct 2021 Evgeny Osipov, Sachin Kahawala, Dilantha Haputhanthri, Thimal Kempitiya, Daswin De Silva, Damminda Alahakoon, Denis Kleyko

Motivated by recent innovations in biologically-inspired neuromorphic hardware, this article presents a novel unsupervised machine learning algorithm named Hyperseed that draws on the principles of Vector Symbolic Architectures (VSA) for fast learning of a topology preserving feature map of unlabelled data.

Few-Shot Learning Language Identification

On Effects of Compression with Hyperdimensional Computing in Distributed Randomized Neural Networks

no code implementations17 Jun 2021 Antonello Rosato, Massimo Panella, Evgeny Osipov, Denis Kleyko

A change of the prevalent supervised learning techniques is foreseeable in the near future: from the complex, computational expensive algorithms to more flexible and elementary training ones.

Dimensionality Reduction Quantization

HyperEmbed: Tradeoffs Between Resources and Performance in NLP Tasks with Hyperdimensional Computing enabled Embedding of n-gram Statistics

no code implementations3 Mar 2020 Pedro Alonso, Kumar Shridhar, Denis Kleyko, Evgeny Osipov, Marcus Liwicki

The embedding achieved on par F1 scores while decreasing the time and memory requirements by several times compared to the conventional n-gram statistics, e. g., for one of the classifiers on a small dataset, the memory reduction was 6. 18 times; while train and test speed-ups were 4. 62 and 3. 84 times, respectively.

Density Encoding Enables Resource-Efficient Randomly Connected Neural Networks

3 code implementations19 Sep 2019 Denis Kleyko, Mansour Kheffache, E. Paxon Frady, Urban Wiklund, Evgeny Osipov

The deployment of machine learning algorithms on resource-constrained edge devices is an important challenge from both theoretical and applied points of view.

BIG-bench Machine Learning

Integer Echo State Networks: Efficient Reservoir Computing for Digital Hardware

no code implementations1 Jun 2017 Denis Kleyko, E. Paxon Frady, Mansour Kheffache, Evgeny Osipov

We propose an approximation of Echo State Networks (ESN) that can be efficiently implemented on digital hardware based on the mathematics of hyperdimensional computing.

Computational Efficiency Time Series +1

Autoscaling Bloom Filter: Controlling Trade-off Between True and False Positives

1 code implementation10 May 2017 Denis Kleyko, Abbas Rahimi, Ross W. Gayler, Evgeny Osipov

A Bloom filter is a simple data structure supporting membership queries on a set.

Data Structures and Algorithms

Holographic Graph Neuron: a Bio-Inspired Architecture for Pattern Processing

no code implementations15 Jan 2015 Denis Kleyko, Evgeny Osipov, Alexander Senior, Asad I. Khan, Y. Ahmet Şekercioğlu

This article proposes the use of Vector Symbolic Architectures for implementing Hierarchical Graph Neuron, an architecture for memorizing patterns of generic sensor stimuli.

Cannot find the paper you are looking for? You can Submit a new open access paper.