Related papers: Learning to Shard: RL for Co-optimizing the Parall…

In the rapidly evolving research on artificial intelligence (AI) the demand for fast, computationally efficient, and scalable solutions has increased in recent years. The problem of optimizing the computing resources for distributed machine…

Federated learning systems have been identified as an efficient approach to scaling distributed model training with a large amount of participants or data owners while guaranteeing data privacy. To apply the current most popular pre-trained…

Distributed Computation has been a recent trend in engineering research. Parallel Computation is widely used in different areas of Data Mining, Image Processing, Simulating Models, Aerodynamics and so forth. One of the major usage of…

Large deep learning models have shown great potential for delivering exceptional results in various applications. However, the training process can be incredibly challenging due to the models' vast parameter sizes, often consisting of…

In this paper we consider the problem of identifying intersections between two sets of d-dimensional axis-parallel rectangles. This is a common problem that arises in many agent-based simulation studies, and is of central importance in the…

Multi-task representation learning is an approach that learns shared latent representations across related tasks, facilitating knowledge transfer and improving sample efficiency. This paper introduces a novel approach to multi-task…

Reinforcement learning (RL) has become the core post-training technique for large language models (LLMs). RL for LLMs involves two stages: generation and training. The LLM first generates samples online, which are then used to derive…

The recent Natural Language Processing techniques have been refreshing the state-of-the-art performance at an incredible speed. Training huge language models is therefore an imperative demand in both industry and academy. However, huge…

State-of-the-art LLMs are powered by scaling -- scaling model size, dataset size and cluster size. It is economically infeasible to extensively tune hyperparameter for the largest runs. Instead, approximately optimal hyperparameters must be…

Scaling reinforcement learning (RL) has shown strong promise for enhancing the reasoning abilities of large language models (LLMs), particularly in tasks requiring long chain-of-thought generation. However, RL training efficiency is often…

The remarkable capabilities of Large Language Models (LLMs) are overshadowed by their immense computational cost. While recent work has shown that many LLM layers can be reordered or even removed with minimal impact on accuracy, these…

As large language models (LLMs) become widespread in various application domains, a critical challenge the AI community is facing is how to train these large AI models in a cost-effective manner. Existing LLM training plans typically employ…

In large scale machine learning and data mining problems with high feature dimensionality, the Euclidean distance between data points can be uninformative, and Distance Metric Learning (DML) is often desired to learn a proper similarity…

Large language models are often adapted using parameter-efficient techniques such as Low-Rank Adaptation (LoRA), formulated as $y = W_0x + BAx$, where $W_0$ is the pre-trained parameters and $x$ is the input to the adapted layer. While…

Deep learning models trained on large data sets have been widely successful in both vision and language domains. As state-of-the-art deep learning architectures have continued to grow in parameter count so have the compute budgets and times…

Modern RL-based post-training for large language models (LLMs) co-locate trajectory sampling and policy optimisation on the same GPU cluster, forcing the system to switch between inference and training workloads. This serial context…

Techniques such as ensembling and distillation promise model quality improvements when paired with almost any base model. However, due to increased test-time cost (for ensembles) and increased complexity of the training pipeline (for…

Computing a Single-Linkage Dendrogram (SLD) is a key step in the classic single-linkage hierarchical clustering algorithm. Given an input edge-weighted tree $T$, the SLD of $T$ is a binary dendrogram that summarizes the $n-1$ clusterings…

Existing works on large language model (LLM) decomposition mainly focus on improving performance on downstream tasks, but they ignore the poor parallel inference performance when trying to scale up the model size. To mitigate this important…

We introduce a new sorting algorithm that is the combination of ML-enhanced sorting with the In-place Super Scalar Sample Sort (IPS4o). The main contribution of our work is to achieve parallel ML-enhanced sorting, as previous algorithms…