Picture for Geethan Karunaratne

Geethan Karunaratne

IBM Research - Zurich

On the Role of Noise in Factorizers for Disentangling Distributed Representations

Add code
Nov 30, 2024
Viaarxiv icon

Limits of Transformer Language Models on Learning Algorithmic Compositions

Add code
Feb 13, 2024
Viaarxiv icon

Zero-shot Classification using Hyperdimensional Computing

Add code
Jan 30, 2024
Viaarxiv icon

TCNCA: Temporal Convolution Network with Chunked Attention for Scalable Sequence Processing

Add code
Dec 09, 2023
Viaarxiv icon

MIMONets: Multiple-Input-Multiple-Output Neural Networks Exploiting Computation in Superposition

Add code
Dec 05, 2023
Viaarxiv icon

Factorizers for Distributed Sparse Block Codes

Add code
Mar 24, 2023
Viaarxiv icon

In-memory factorization of holographic perceptual representations

Add code
Nov 09, 2022
Viaarxiv icon

In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory

Add code
Jul 14, 2022
Figure 1 for In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory
Figure 2 for In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory
Figure 3 for In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory
Figure 4 for In-memory Realization of In-situ Few-shot Continual Learning with a Dynamically Evolving Explicit Memory
Viaarxiv icon

Constrained Few-shot Class-incremental Learning

Add code
Mar 30, 2022
Figure 1 for Constrained Few-shot Class-incremental Learning
Figure 2 for Constrained Few-shot Class-incremental Learning
Figure 3 for Constrained Few-shot Class-incremental Learning
Figure 4 for Constrained Few-shot Class-incremental Learning
Viaarxiv icon

Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks

Add code
Mar 11, 2022
Figure 1 for Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks
Figure 2 for Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks
Figure 3 for Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks
Figure 4 for Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks
Viaarxiv icon