activity
20242026
collaborators
Showing cs.LGShow all

6 papers · 1 filter

cs.LG2025

Why Are Positional Encodings Nonessential for Deep Autoregressive Transformers? Revisiting a Petroglyph

Kazuki Irie

Do autoregressive Transformer language models require explicit positional encodings (PEs)? The answer is 'no' provided they have more than one layer -- they can distinguish sequenc…

cs.LG2025

Metalearning Continual Learning Algorithms

Kazuki Irie, Róbert Csordás, Jürgen Schmidhuber

General-purpose learning systems should improve themselves in open-ended fashion in ever-changing environments. Conventional learning algorithms for neural networks, however, suffe…

cs.LG2024

Dissecting the Interplay of Attention Paths in a Statistical Mechanics Theory of Transformers

Lorenzo Tiberi, Francesca Mignacco, Kazuki Irie +1

Despite the remarkable empirical performance of Transformers, their theoretical understanding remains elusive. Here, we consider a deep multi-head self-attention network, that is c…

cs.LG2024

MoEUT: Mixture-of-Experts Universal Transformers

Róbert Csordás, Kazuki Irie, Jürgen Schmidhuber +2

Previous work on Universal Transformers (UTs) has demonstrated the importance of parameter sharing across layers. By allowing recurrence in depth, UTs have advantages over standard…

cs.LG2024

SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention

Róbert Csordás, Piotr Piękos, Kazuki Irie +1

Despite many recent works on Mixture of Experts (MoEs) for resource-efficient Transformer language models, existing methods mostly focus on MoEs for feedforward layers. Previous at…

cs.LG2024

Self-Organising Neural Discrete Representation Learning à la Kohonen

Kazuki Irie, Róbert Csordás, Jürgen Schmidhuber

Unsupervised learning of discrete representations in neural networks (NNs) from continuous ones is essential for many modern applications. Vector Quantisation (VQ) has become popul…