Pentti Kanerva

Timeline

1. 1956–1967

   Education and computing in Finland

   He earns an A.A. from Warren Wilson College in 1956 and an M.S. in forestry, with a minor in mathematics and statistics, from the University of Helsinki in 1964. From 1961 to 1967, he works in statistics and programming at the Forest Research Institute of Finland, the Finnish State Computer Center and the University of Tampere.

2. 1967–1985

   Stanford systems and a PhD

   In 1967, he moves from Finland to Stanford to work in Patrick Suppes' computer laboratory, where he designs and builds systems for computer-assisted instruction. He serves as a systems programmer and research assistant at Stanford's Institute for Mathematical Studies in the Social Sciences through 1978, then as a computer systems specialist at its Center for Information Technology from 1979 to 1983. In 1984, he earns a philosophy PhD at Stanford and begins a postdoctoral fellowship at the Center for the Study of Language and Information through 1985.

3. 1985–1992

   Sparse Distributed Memory at NASA

   He serves as Senior Scientist at the Research Institute for Advanced Computer Science at NASA Ames from 1985 to 1992. In 1988, he publishes Sparse Distributed Memory, a book based on his Stanford philosophy PhD thesis and a mathematical model of human long-term memory.

4. 1993–2002

   Random indexing at SICS

   He serves as a Senior Researcher at the Swedish Institute of Computer Science from 1993 to 2002. In 2000, he coauthors Random Indexing of Text Samples for Latent Semantic Analysis with Jan Kristoferson and Anders Holst, a paper on using random indexing to learn semantic relations from language.

5. 2002–2006

   Research at Redwood Institute

   He serves as a Senior Researcher at Redwood Neuroscience Institute in 2002–03 and as a Research Affiliate from 2003 to 2006.

6. 2009

   Hyperdimensional computing

   In 2009, he publishes “Hyperdimensional Computing: An Introduction to Computing in Distributed Representation with High-Dimensional Random Vectors” in Cognitive Computation. He uses random distributed representations in spaces with thousands of dimensions to model the brain’s code and computing.

7. 2025

   Autonomous learning architecture

   In 2025, he posts “Autonomous Learning with High-Dimensional Computing Architecture Similar to von Neumann's.” The paper models human and animal learning with high-dimensional vectors and includes a high-capacity vector memory.

8. As of October 2026

   Research scientist at Berkeley

   He is a Research Scientist at UC Berkeley's Redwood Center for Theoretical Neuroscience.