arXiv research

QMIX combines local values to train decentralised policies in a centralised setting.

We refine Cohen and Lustig's description of centralisers of Dehn twists of free groups. We show that the centraliser of a Dehn twist of a free group has a subgroup of finite index that has a finite classifying space. We describe an algorithm to find a presentation of the centraliser. We use this algorithm to give an ex…

QMIX combines per-agent values to create decentralised policies.

FACMAC combines deep policy gradients with factored critic for multi-agent reinforcement learning.

The virtually cyclic dimension of Out(F_N) is finite and related properties are established.

The paper studies algebraic structures related to quantum groups.

Study of automorphisms and splittings of special groups, showing infinite groups under certain conditions.

The thesis explores centralisers and Hecke algebras in representation theory with applications to knots and physics.

This paper introduces a new metric to improve the performance of AMMs over centralised exchanges.

We introduce the palindromic automorphism group and the palindromic Torelli group of a right-angled Artin group A_G. The palindromic automorphism group Pi A_G is related to the principal congruence subgroups of GL(n,Z) and to the hyperelliptic mapping class group of an oriented surface, and sits inside the centraliser …

The paper presents methods to write presentations for Dehn quandles.

The paper studies the action of automorphisms on train tracks and finds that the set of minimally displaced points is co-compact.

Paper shows federated learning can train models on private data.

DistGP models multi-robot mapping with distributed Gaussian process learning.

The paper proves a conjecture about the dimensions of centralizer algebras related to quantum super-algebras.

Study growth rates of automorphisms of special groups.

FTX's failure linked to Terra-Luna collapse and Binance's influence.

New post-processing methods improve word embedding performance.

Graph-dependent implicit regularisation improves Distributed SGD for convex problems.

We describe a new class of holonomy groups on pseudo-Riemannian manifolds. Namely, we prove the following theorem. Let g be a nondegenerate bilinear form on a vector space V, and L:V -> V a g-symmetric operator. Then the identity component of the centraliser of L in SO(g) is a holonomy group for a suitable Levi-Civita …

Deep RL agent secures 2nd place in CityLearn Challenge for district demand management.

This work uses reinforcement learning to optimize task scheduling and execution in a dynamic multi-agent warehouse environment.

MAVEN improves multi-agent exploration by hybridizing value and policy-based methods.

Finite order elements with infinite centralizers in 3-manifold groups imply specific structure.

This work makes federated Bayesian learning differentially private.

A modular GP framework for efficient transfer learning.

FedSLIM optimizes compact pattern models across distributed databases without sharing raw data.

A framework for multi-agent learning improves coordination through a memory-driven communication protocol.

The article provides formulas to hedge impermanent loss in decentralized markets.

Fix a simple complex Lie group G and a principal sl(2,C) subalgebra of Lie(G). Then the moduli space of semi-stable, topologically trivial G-Higgs bundles on a hyperbolic, spin Riemann surface acquires a marked point. This is the unique C*-fixed point on the Hitchin section. We describe a universal analytic family of d…

EMIX minimizes surprise in multi-agent reinforcement learning.

Paper predicts transaction confirmation time in Ethereum blockchain using machine learning.

We provide two distributed confidence ball algorithms for solving linear bandit problems in peer to peer networks with limited communication capabilities. For the first, we assume that all the peers are solving the same linear bandit problem, and prove that our algorithm achieves the optimal asymptotic regret rate of a…

Let t_a be the Dehn twist about a circle a on an orientable surface. It is well known that for each circle b and an integer n, I(t_a^n(b),b)=|n|I(a,b)^2, where I(,) is the geometric intersection number. We prove a similar formula for circles on nonorientable surfaces. As a corollary we prove some algebraic properties o…

Garside groupoids, as recently introduced by Krammer, generalise Garside groups. A weak Garside group is a group that is equivalent as a category to a Garside groupoid. We show that any periodic loop in a Garside groupoid $\CG$ may be viewed as a Garside element for a certain Garside structure on another Garside groupo…

A novel hierarchical Bayesian approach to Federated Learning reduces data exposure and improves convergence rates.

Optimizes trading in CFMMs and exchanges using deep learning.

Study on conformal transformations of Cahen-Wallach spaces, focusing on fixed points and discontinuous groups.

Paper proves equivariant Fried conjecture for specific flows.

This work benchmarks MARL algorithms in cooperative tasks.

We consider linear groups which do not contain unipotent elements of infinite order, which includes all linear groups in positive characteristic, and show that this class of groups has good properties which resemble those held by groups of non positive curvature and which do not hold for arbitrary characteristic zero l…

Generators found for automorphisms of special groups.

The paper optimizes exceptions in a statistical production system using machine learning.

Parallelizes graph embedding for large graphs.

This study optimizes trading and arbitrage in decentralized finance's CPMs, revealing convexity costs and developing efficient strategies.

Improved fraud detection in finance with quantum-enhanced federated learning.

OpenAlpha validates decentralized capital strategies using game theory and market aggregation.

We study the problem of determining the isomorphism classes of the virtually cyclic subgroups of the n-string braid groups B_n(S^2) of the 2-sphere S^2. If n is odd, or if n is even and sufficiently large, we obtain the complete classification. For small even values of n, the classification is complete up to an explici…