13 papers
Safety without alignment
András Kornai, Michael Bukatin, Zsolt Zombori
Currently, the dominant paradigm in AI safety is alignment with human values. Here we describe progress on developing an alternative approach to safety, based on ethical rationalis…
Dataflow Matrix Machines and V-values: a Bridge between Programs and Neural Nets
Michael Bukatin, Jon Anthony
1) Dataflow matrix machines (DMMs) generalize neural nets by replacing streams of numbers with linear streams (streams supporting linear combinations), allowing arbitrary input and…
Dataflow Matrix Machines as a Model of Computations with Linear Streams
Michael Bukatin, Jon Anthony
We overview dataflow matrix machines as a Turing complete generalization of recurrent neural networks and as a programming platform. We describe vector space of finite prefix trees…
Notes on Pure Dataflow Matrix Machines: Programming with Self-referential Matrix Transformations
Michael Bukatin, Steve Matthews, Andrey Radul
Dataflow matrix machines are self-referential generalized recurrent neural nets. The self-referential mechanism is provided via a stream of matrices defining the connectivity and w…
Revisiting EPRL: All Finite-Dimensional Solutions by Naimark's Fundamental Theorem
Leonid Perlov, Michael Bukatin
In this paper we research all possible finite-dimensional representations and corresponding values of the Barbero-Immirzi parameter contained in EPRL simplicity constraints by usin…
Programming Patterns in Dataflow Matrix Machines and Generalized Recurrent Neural Nets
Michael Bukatin, Steve Matthews, Andrey Radul
Dataflow matrix machines arise naturally in the context of synchronous dataflow programming with linear streams. They can be viewed as a rather powerful generalization of recurrent…
Dataflow matrix machines as programmable, dynamically expandable, self-referential generalized recurrent neural networks
Michael Bukatin, Steve Matthews, Andrey Radul
Dataflow matrix machines are a powerful generalization of recurrent neural networks. They work with multiple types of linear streams and multiple types of neurons, including higher…
Dataflow Matrix Machines as a Generalization of Recurrent Neural Networks
Michael Bukatin, Steve Matthews, Andrey Radul
Dataflow matrix machines are a powerful generalization of recurrent neural networks. They work with multiple types of arbitrary linear streams, multiple types of powerful neurons,…
Dataflow Graphs as Matrices and Programming with Higher-order Matrix Elements
Michael Bukatin, Steve Matthews
We consider dataflow architecture for two classes of computations which admit taking linear combinations of execution runs: probabilistic sampling and generalized animation. We imp…
Almost Continuous Transformations of Software and Higher-order Dataflow Programming
Michael Bukatin, Steve Matthews
We consider two classes of stream-based computations which admit taking linear combinations of execution runs: probabilistic sampling and generalized animation. The dataflow archit…
Linear Models of Computation and Program Learning
Michael Bukatin, Steve Matthews
We consider two classes of computations which admit taking linear combinations of execution runs: probabilistic sampling and generalized animation. We argue that the task of progra…
Mathematics of Domains
Michael A. Bukatin
Two groups of naturally arising questions in the mathematical theory of domains for denotational semantics are addressed. Domains are equipped with Scott topology and represent dat…
Barbero-Immirzi parameter as a solution of the simplicity constraints
Leonid Perlov, Michael Bukatin
In this paper we naturally obtain the values of the Barbero-Immirzi parameter as the solution of the simplicity constraints rather than setting it a priori. Particularly the Main t…