This page presents the recordings of the seminar course on HD/VSA titled "Computing with High-Dimensional Vectors". The course was organized by Redwood Center for Theoretical Neuroscience and Berkeley Wireless Research Center at UC Berkeley during fall 2021. The course consisted of twelve modules. A lecture within each module was given by one of the instructors or an invited lecturer.
Module1_Lecture_Olshausen.MP4Module1_Lecture_Kanerva.mp4Module 1: Introduction to Computing with High-Dimensional Vectors
Speakers: Bruno Olshausen (Part 1; slides) & Pentti Kanerva (Part 2; slides)
Date: September 1, 2021
Module's focus paper: Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random Vectors
Module2_Overview-of-different-HD-Computing-VSA-models_Kleyko.mp4Module 2: Overview of Different HD Computing/VSA Models
Speaker: Denis Kleyko (slides)
Date: September 8, 2021
Module's focus paper: A Comparison of Vector Symbolic Architectures
Module3_Word-embeddings-with-HD-Computing-VSA_Moughan.mp4Module 3: Semantic Vectors
Speaker: Ryan Moughan (slides)
Date: September 15, 2021
Module's focus paper: Representing Word Meaning and Order Information in a Composite Holographic Lexicon
Module4_Representation-and-manipulation-of-data-structures_Kleyko.mp4Module 4: Data structures
Speaker: Denis Kleyko (slides)
Date: September 22, 2021
Module's focus paper: Vector Symbolic Architectures as a Computing Framework for Nanoscale Hardware
Module5_Solving-factorization-problems-with-VSAs-and-resonator-networks_Frady.mp4Module 5: Resonator networks
Speaker: E. Paxon Frady (slides)
Date: September 29, 2021
Module's focus papers: Resonator Networks, 1: An Efficient Solution for Factoring High-Dimensional, Distributed Representations of Data Structures
RG_lecture_zoom.mp4Module 6: Analogical reasoning
Speaker: Ross Gayler (slides)
Date: October 6, 2021
Module's focus paper: A Distributed Basis for Analogical Mapping
Module7_Sommer_Capacity_of_VSA_representations.mp4Module 7: Connections to Information Theory
Speaker: Fritz Sommer (slides)
Date: October 13, 2021
Module's focus paper: A Theory of Sequence Indexing and Working Memory in Recurrent Neural Network
Module8_Kymn_Locality-preserving-encodings.mp4Module 8: Locality-Preserving Encodings: Representing Continuous Values and Functions
Speaker: Chris Kymn (slides)
Date: October 20, 2021
Module's focus paper: Computing on Functions Using Randomized Vector Representations
Module9_Galindez-Olascoaga_Solving-classification-problems.mp4Module 9: Solving Classification Problems
Speaker: Laura Galindez Olascoaga (slides)
Date: October 27, 2021
Module's focus paper: Efficient Biosignal Processing Using Hyperdimensional Computing: Network Templates for Combined Learning and Classification of ExG Signals
Module10_Kleyko_Relations-to-neural-networks.mp4Module 10: Relations to Neural Networks
Speaker: Denis Kleyko (slides)
Date: November 3, 2021
Note: Due to a technical problem approximately one minute of the recording is missing at 00:31:55.
Module's focus paper: Associative Long Short-Term Memory
Module11_Ibrahim_Efficient-hardware-design-for-hyperdimensional-learning.mp4Module 11: Hardware Implementations
Speaker: Mohamed Ibrahim (slides)
Date: November 10, 2021
Module's focus paper: In-Memory Hyperdimensional Computing
Module12_Huang_Communication.MP4Module 12: Communications
Speaker: Ping-Chen Huang (slides)
Date: November 17, 2021
Module's focus paper: HDM: Hyper-Dimensional Modulation for Robust Low-Power Communications