arXiv subjects

John Donaghy

Publications and source records attributed to John Donaghy.

5 recordsLinked to original sources

arXiv2026

DEI: Diversity in Evolutionary Inference for Quality-Diversity Search

We present DEI: Diversity in Evolutionary Inference, a distributed Quality-Diversity (QD) search framework that assigns heterogeneous large language models (LLMs) as mutation operators across peer nodes communicating with non-blocking collective operations. Unlike homogeneous parallel search, which replicates a single model's inductive biases across all workers, DEI treats each LLM's distinct creative prior as a complementary source of behavioral novelty. Extending the Digital Red Queen framework with DEI, nodes share local optimal solutions at the end of each round to seed the next round's population. This creates cross-model adversarial pressure that drives robustness beyond intra-model self-play. Evaluated on the Core War domain, a competitive programming benchmark in which Redcode warrior programs battle inside a simulated machine, a four-node heterogeneous ensemble (GPT-5.4-mini, Claude Sonnet 4.6, GPT-5.2, and Claude Haiku 4.5) achieves 124 percent higher merged-archive QD-Score (45.90 vs. 20.46) and 28 percent higher coverage (80.6 percent vs. 63.0 percent of cells) than a single-node baseline at equal total LLM-call budget. The heterogeneous ensemble also outperforms an equally-budgeted homogeneous ensemble on QD-Score, coverage, and held-out solution generality across all four model families. These results provide the first empirical evidence that model diversity, not merely parallelism, is the key driver of gain in distributed LLM-based QD search.

cs.LG↗

arXiv2025

Sharing is Caring: Efficient LM Post-Training with Collective RL Experience Sharing

Post-training language models (LMs) with reinforcement learning (RL) can enhance their complex reasoning capabilities without supervised fine-tuning, as demonstrated by DeepSeek-R1-Zero. However, effectively utilizing RL for LMs requires significant parallelization to scale-up inference, which introduces non-trivial technical challenges (e.g. latency, memory, and reliability) alongside ever-growing financial costs. We present Swarm sAmpling Policy Optimization (SAPO), a fully decentralized and asynchronous RL post-training algorithm. SAPO is designed for decentralized networks of heterogenous compute nodes, where each node manages its own policy model(s) while "sharing" rollouts with others in the network; no explicit assumptions about latency, model homogeneity, or hardware are required and nodes can operate in silo if desired. As a result, the algorithm avoids common bottlenecks in scaling RL post-training while also allowing (and even encouraging) new possibilities. By sampling rollouts "shared" across the network, it enables "Aha moments" to propagate, thereby bootstrapping the learning process. In this paper we show SAPO achieved cumulative reward gains of up to 94% in controlled experiments. We also share insights from tests on a network with thousands of nodes contributed by Gensyn community members running the algorithm on diverse hardware and models during an open-source demo.

cs.LG↗

arXiv2025

NoLoCo: No-all-reduce Low Communication Training Method for Large Models

Training large language models is generally done on clusters containing thousands of accelerators, communicating over a high-bandwidth interconnect. Scaling up these clusters is expensive and can become impractical, imposing limits on the size of models that can be trained. Several recent studies have proposed training methods that are less communication intensive, avoiding the need for compute clusters with extremely high interconnect speeds. These low communication training methods still employ a global synchronization step for model parameters, which can be too costly with a high number of participants, as the communication cost scales quadratically with group size. In this work, we propose a novel optimization method, NoLoCo, that does not explicitly synchronize all model parameters during training and does not require any collective communication. NoLoCo implicitly synchronizes model weights via a novel variant of the Nesterov momentum optimizer by partially averaging model weights within randomly selected subgroups. We provide both a theoretical convergence analysis of our optimizer and empirical results from language model training. Our method requires significantly less communication than fully sharded data parallel training and DiLoCo, a widely used low-communication baseline. Moreover, our method avoids global blocking communication, thereby reducing accelerator idle time. Our experiments show that NoLoCo is more communication-efficient than DiLoCo, improving final perplexity by up to $4\%$ and converging up to $4\times$ faster in wall-clock time across a range of worker counts, model sizes, and communication bandwidths.

cs.LG↗

arXiv2023

In Search of a Data Driven Symbolic Multi-Fluid 10-Moment Model Closure

The inclusion of kinetic effects into fluid models has been a long standing problem in magnetic reconnection and plasma physics. Generally the pressure tensor is reduced to a scalar which is an approximation used to aid in the modeling of large scale global systems such as the Earth's magnetosphere. This unfortunately omits important kinetic physics which have been shown to play a crucial role in collisionless regimes. The multi-fluid 10-moment model on the other-hand retains the full symmetric pressure tensor. The 10-moment model is constructed by taking moments of the Vlasov equation up to second order, and includes the scalar density, the vector bulk-flow, and the symmetric pressure tensor for a total of 10 separate components. Use of the multi-fluid 10-moment model requires a closure which truncates the cascading system of equations. Here we look to leverage data-driven methodologies to seek a closure which may improve physical fidelity of the 10-moment multi-fluid model in collisionless regimes. Specifically we use the Sparse Identification of Nonlinear Dynamics (SINDy) method for symbolic equation discovery to seek the truncating closure from fully kinetic particle-in-cell simulation data, which inherently retains the relevant kinetic physics. We verify our method by reproducing the 10-moment model from the PIC particle data and use the method to generate a closure truncating the 10-moment model which is analyzed through the nonlinear phase of reconnection.

physics.plasm-ph↗

arXiv2021

Inference and De-Noising of Non-Gaussian Particle Distribution Functions: A Generative Modeling Approach

The particle-in-cell numerical method of plasma physics balances a trade-off between computational cost and intrinsic noise. Inference on data produced by these simulations generally consists of binning the data to recover the particle distribution function, from which physical processes may be investigated. In addition to containing noise, the distribution function is temporally dynamic and can be non-gaussian and multi-modal, making the task of modeling it difficult. Here we demonstrate the use of normalizing flows to learn a smooth, tractable approximation to the noisy particle distribution function. We demonstrate that the resulting data driven likelihood conserves relevant physics and may be extended to encapsulate the temporal evolution of the distribution function.

physics.plasm-ph↗