Multi-agent reinforcement learning (MARL) algorithms often struggle to find strategies close to Pareto optimal Nash Equilibrium, owing largely to the lack of efficient exploration. The problem is exacerbated in sparse-reward settings, caused by the larger variance exhibited in policy learning. This paper introduces MESA, a novel meta-exploration method for cooperative multi-agent learning. It learns to explore by first identifying the agents' high-rewarding joint state-action subspace from training tasks and then learning a set of diverse exploration policies to "cover" the subspace. These trained exploration policies can be integrated with any off-policy MARL algorithm for test-time tasks. We first showcase MESA's advantage in a multi-step matrix game. Furthermore, experiments show that with learned exploration policies, MESA achieves significantly better performance in sparse-reward tasks in several multi-agent particle environments and multi-agent MuJoCo environments, and exhibits the ability to generalize to more challenging tasks at test time.
相關內容
Efficient finetuning of large language models (LLMs) aims to adapt the LLMs with reduced computation and memory cost. Previous LoRA-based approaches initialize the low-rank matrices with gaussian distribution and zero values, while keeping the original weight matrices frozen. However, the trainable model parameters optimized in an unguided subspace might have interference with the well-learned subspace of the pretrained weight matrix. In this paper, we propose MiLoRA, a simple yet effective LLM finetuning approach that only updates the minor singular components of the weight matrix while keeping the principle singular components frozen. It is observed that the minor matrix corresponds to the noisy or long-tail information, while the principle matrix contains important knowledge. The MiLoRA initializes the low-rank matrices within a subspace that is orthogonal to the principle matrix, thus the pretrained knowledge is expected to be well preserved. During finetuning, MiLoRA makes the most use of the less-optimized subspace for learning the finetuning dataset. Extensive experiments on commonsense reasoning, math reasoning and instruction following benchmarks present the superior performance of our method.
This study investigates the application of advanced machine learning models, specifically Long Short-Term Memory (LSTM) networks and Gradient Booster models, for accurate energy consumption estimation within a Kubernetes cluster environment. It aims to enhance sustainable computing practices by providing precise predictions of energy usage across various computing nodes. Through meticulous analysis of model performance on both master and worker nodes, the research reveals the strengths and potential applications of these models in promoting energy efficiency. The LSTM model demonstrates remarkable predictive accuracy, particularly in capturing dynamic computing workloads over time, evidenced by low mean squared error (MSE) rates and the ability to closely track actual energy consumption trends. Conversely, the Gradient Booster model showcases robustness and adaptability across different computational environments, despite slightly higher MSE values. The study underscores the complementary nature of these models in advancing sustainable computing practices, suggesting their integration into energy management systems could significantly enhance environmental sustainability in technology operations.
In-context learning (ICL) facilitates Large Language Models (LLMs) exhibiting emergent ability on downstream tasks without updating billions of parameters. However, in the area of multi-modal Large Language Models (MLLMs), two problems hinder the application of multi-modal ICL: (1) Most primary MLLMs are only trained on single-image datasets, making them unable to read multi-modal demonstrations. (2) With the demonstrations increasing, thousands of visual tokens highly challenge hardware and degrade ICL performance. During preliminary explorations, we discovered that the inner LLM tends to focus more on the linguistic modality within multi-modal demonstrations to generate responses. Therefore, we propose a general and light-weighted framework \textbf{AIM} to tackle the mentioned problems through \textbf{A}ggregating \textbf{I}mage information of \textbf{M}ultimodal demonstrations to the dense latent space of the corresponding linguistic part. Specifically, AIM first uses the frozen backbone MLLM to read each image-text demonstration and extracts the vector representations on top of the text. These vectors naturally fuse the information of the image-text pair, and AIM transforms them into fused virtual tokens acceptable for the inner LLM via a trainable projection layer. Ultimately, these fused tokens function as variants of multi-modal demonstrations, fed into the MLLM to direct its response to the current query as usual. Because these fused tokens stem from the textual component of the image-text pair, a multi-modal demonstration is nearly reduced to a pure textual demonstration, thus seamlessly applying to any MLLMs. With its de facto MLLM frozen, AIM is parameter-efficient and we train it on public multi-modal web corpora which have nothing to do with downstream test tasks.
Graph Neural Networks (GNNs) have superior capability in learning graph data. Full-graph GNN training generally has high accuracy, however, it suffers from large peak memory usage and encounters the Out-of-Memory problem when handling large graphs. To address this memory problem, a popular solution is mini-batch GNN training. However, mini-batch GNN training increases the training variance and sacrifices the model accuracy. In this paper, we propose a new memory-efficient GNN training method using spanning subgraph, called SpanGNN. SpanGNN trains GNN models over a sequence of spanning subgraphs, which are constructed from empty structure. To overcome the excessive peak memory consumption problem, SpanGNN selects a set of edges from the original graph to incrementally update the spanning subgraph between every epoch. To ensure the model accuracy, we introduce two types of edge sampling strategies (i.e., variance-reduced and noise-reduced), and help SpanGNN select high-quality edges for the GNN learning. We conduct experiments with SpanGNN on widely used datasets, demonstrating SpanGNN's advantages in the model performance and low peak memory usage.
On-device continual learning (CL) requires the co-optimization of model accuracy and resource efficiency to be practical. This is extremely challenging because it must preserve accuracy while learning new tasks with continuously drifting data and maintain both high energy and memory efficiency to be deployable on real-world devices. Typically, a CL method leverages one of two types of backbone networks: CNN or ViT. It is commonly believed that CNN-based CL excels in resource efficiency, whereas ViT-based CL is superior in model performance, making each option attractive only for a single aspect. In this paper, we revisit this comparison while embracing powerful pre-trained ViT models of various sizes, including ViT-Ti (5.8M parameters). Our detailed analysis reveals that many practical options exist today for making ViT-based methods more suitable for on-device CL, even when accuracy, energy, and memory are all considered. To further expand this impact, we introduce REP, which improves resource efficiency specifically targeting prompt-based rehearsal-free methods. Our key focus is on avoiding catastrophic trade-offs with accuracy while trimming computational and memory costs throughout the training process. We achieve this by exploiting swift prompt selection that enhances input data using a carefully provisioned model, and by developing two novel algorithms-adaptive token merging (AToM) and adaptive layer dropping (ALD)-that optimize the prompt updating stage. In particular, AToM and ALD perform selective skipping across the data and model-layer dimensions without compromising task-specific features in vision transformer models. Extensive experiments on three image classification datasets validate REP's superior resource efficiency over current state-of-the-art methods.
Graph Neural Networks (GNNs) have proven highly effective in various machine learning (ML) tasks involving graphs, such as node/graph classification and link prediction. However, explaining the decisions made by GNNs poses challenges because of the aggregated relational information based on graph structure, leading to complex data transformations. Existing methods for explaining GNNs often face limitations in systematically exploring diverse substructures and evaluating results in the absence of ground truths. To address this gap, we introduce GNNAnatomy, a model- and dataset-agnostic visual analytics system designed to facilitate the generation and evaluation of multi-level explanations for GNNs. In GNNAnatomy, we employ graphlets to elucidate GNN behavior in graph-level classification tasks. By analyzing the associations between GNN classifications and graphlet frequencies, we formulate hypothesized factual and counterfactual explanations. To validate a hypothesized graphlet explanation, we introduce two metrics: (1) the correlation between its frequency and the classification confidence, and (2) the change in classification confidence after removing this substructure from the original graph. To demonstrate the effectiveness of GNNAnatomy, we conduct case studies on both real-world and synthetic graph datasets from various domains. Additionally, we qualitatively compare GNNAnatomy with a state-of-the-art GNN explainer, demonstrating the utility and versatility of our design.
We present ABEX, a novel and effective generative data augmentation methodology for low-resource Natural Language Understanding (NLU) tasks. ABEX is based on ABstract-and-EXpand, a novel paradigm for generating diverse forms of an input document -- we first convert a document into its concise, abstract description and then generate new documents based on expanding the resultant abstraction. To learn the task of expanding abstract descriptions, we first train BART on a large-scale synthetic dataset with abstract-document pairs. Next, to generate abstract descriptions for a document, we propose a simple, controllable, and training-free method based on editing AMR graphs. ABEX brings the best of both worlds: by expanding from abstract representations, it preserves the original semantic properties of the documents, like style and meaning, thereby maintaining alignment with the original label and data distribution. At the same time, the fundamental process of elaborating on abstract descriptions facilitates diverse generations. We demonstrate the effectiveness of ABEX on 4 NLU tasks spanning 12 datasets and 4 low-resource settings. ABEX outperforms all our baselines qualitatively with improvements of 0.04% - 38.8%. Qualitatively, ABEX outperforms all prior methods from literature in terms of context and length diversity.
As an effective strategy, data augmentation (DA) alleviates data scarcity scenarios where deep learning techniques may fail. It is widely applied in computer vision then introduced to natural language processing and achieves improvements in many tasks. One of the main focuses of the DA methods is to improve the diversity of training data, thereby helping the model to better generalize to unseen testing data. In this survey, we frame DA methods into three categories based on the diversity of augmented data, including paraphrasing, noising, and sampling. Our paper sets out to analyze DA methods in detail according to the above categories. Further, we also introduce their applications in NLP tasks as well as the challenges.
Meta reinforcement learning (meta-RL) extracts knowledge from previous tasks and achieves fast adaptation to new tasks. Despite recent progress, efficient exploration in meta-RL remains a key challenge in sparse-reward tasks, as it requires quickly finding informative task-relevant experiences in both meta-training and adaptation. To address this challenge, we explicitly model an exploration policy learning problem for meta-RL, which is separated from exploitation policy learning, and introduce a novel empowerment-driven exploration objective, which aims to maximize information gain for task identification. We derive a corresponding intrinsic reward and develop a new off-policy meta-RL framework, which efficiently learns separate context-aware exploration and exploitation policies by sharing the knowledge of task inference. Experimental evaluation shows that our meta-RL method significantly outperforms state-of-the-art baselines on various sparse-reward MuJoCo locomotion tasks and more complex sparse-reward Meta-World tasks.
State-of-the-art Convolutional Neural Network (CNN) benefits a lot from multi-task learning (MTL), which learns multiple related tasks simultaneously to obtain shared or mutually related representations for different tasks. The most widely-used MTL CNN structure is based on an empirical or heuristic split on a specific layer (e.g., the last convolutional layer) to minimize different task-specific losses. However, this heuristic sharing/splitting strategy may be harmful to the final performance of one or multiple tasks. In this paper, we propose a novel CNN structure for MTL, which enables automatic feature fusing at every layer. Specifically, we first concatenate features from different tasks according to their channel dimension, and then formulate the feature fusing problem as discriminative dimensionality reduction. We show that this discriminative dimensionality reduction can be done by 1x1 Convolution, Batch Normalization, and Weight Decay in one CNN, which we refer to as Neural Discriminative Dimensionality Reduction (NDDR). We perform ablation analysis in details for different configurations in training the network. The experiments carried out on different network structures and different task sets demonstrate the promising performance and desirable generalizability of our proposed method.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191