Learning Nash equilibrium (NE) in complex zero-sum games with multi-agent reinforcement learning (MARL) can be extremely computationally expensive. Curriculum learning is an effective way to accelerate learning, but an under-explored dimension for generating a curriculum is the difficulty-to-learn of the subgames -- games induced by starting from a specific state. In this work, we present a novel subgame curriculum learning framework for zero-sum games. It adopts an adaptive initial state distribution by resetting agents to some previously visited states where they can quickly learn to improve performance. Building upon this framework, we derive a subgame selection metric that approximates the squared distance to NE values and further adopt a particle-based state sampler for subgame generation. Integrating these techniques leads to our new algorithm, Subgame Automatic Curriculum Learning (SACL), which is a realization of the subgame curriculum learning framework. SACL can be combined with any MARL algorithm such as MAPPO. Experiments in the particle-world environment and Google Research Football environment show SACL produces much stronger policies than baselines. In the challenging hide-and-seek quadrant environment, SACL produces all four emergent stages and uses only half the samples of MAPPO with self-play. The project website is at //sites.google.com/view/sacl-rl.
相關內容
We study a primal-dual reinforcement learning (RL) algorithm for the online constrained Markov decision processes (CMDP) problem, wherein the agent explores an optimal policy that maximizes return while satisfying constraints. Despite its widespread practical use, the existing theoretical literature on primal-dual RL algorithms for this problem only provides sublinear regret guarantees and fails to ensure convergence to optimal policies. In this paper, we introduce a novel policy gradient primal-dual algorithm with uniform probably approximate correctness (Uniform-PAC) guarantees, simultaneously ensuring convergence to optimal policies, sublinear regret, and polynomial sample complexity for any target accuracy. Notably, this represents the first Uniform-PAC algorithm for the online CMDP problem. In addition to the theoretical guarantees, we empirically demonstrate in a simple CMDP that our algorithm converges to optimal policies, while an existing algorithm exhibits oscillatory performance and constraint violation.
Self-supervised learning (SSL) learns representations by leveraging an auxiliary unsupervised task, such as classifying semantically related samples, e.g. different data augmentations or modalities. Of the many approaches to SSL, contrastive methods, e.g. SimCLR, CLIP and VicREG, have gained attention for learning representations that achieve downstream performance close to that of supervised learning. However, a theoretical understanding of the mechanism behind these methods eludes. We propose a generative latent variable model for the data and show that several families of discriminative self-supervised algorithms, including contrastive methods, approximately induce its latent structure over representations, providing a unifying theoretical framework. We also justify links to mutual information and the use of a projection head. Fitting our model generatively, as SimVE, improves performance over previous VAE methods on common benchmarks (e.g. FashionMNIST, CIFAR10, CelebA), narrows the gap to discriminative methods on _content_ classification and, as our analysis predicts, outperforms them where _style_ information is required, taking a step toward task-agnostic representations.
We introduce \textsc{Pok\'eLLMon}, the first LLM-embodied agent that achieves human-parity performance in tactical battle games, as demonstrated in Pok\'emon battles. The design of \textsc{Pok\'eLLMon} incorporates three key strategies: (i) In-context reinforcement learning that instantly consumes text-based feedback derived from battles to iteratively refine the policy; (ii) Knowledge-augmented generation that retrieves external knowledge to counteract hallucination and enables the agent to act timely and properly; (iii) Consistent action generation to mitigate the \textit{panic switching} phenomenon when the agent faces a powerful opponent and wants to elude the battle. We show that online battles against human demonstrates \textsc{Pok\'eLLMon}'s human-like battle strategies and just-in-time decision making, achieving 49\% of win rate in the Ladder competitions and 56\% of win rate in the invited battles. Our implementation and playable battle logs are available at: \url{//github.com/git-disl/PokeLLMon}.
Gaussian processes (GPs) based methods for solving partial differential equations (PDEs) demonstrate great promise by bridging the gap between the theoretical rigor of traditional numerical algorithms and the flexible design of machine learning solvers. The main bottleneck of GP methods lies in the inversion of a covariance matrix, whose cost grows cubically concerning the size of samples. Drawing inspiration from neural networks, we propose a mini-batch algorithm combined with GPs to solve nonlinear PDEs. A naive deployment of a stochastic gradient descent method for solving PDEs with GPs is challenging, as the objective function in the requisite minimization problem cannot be depicted as the expectation of a finite-dimensional random function. To address this issue, we employ a mini-batch method to the corresponding infinite-dimensional minimization problem over function spaces. The algorithm takes a mini-batch of samples at each step to update the GP model. Thus, the computational cost is allotted to each iteration. Using stability analysis and convexity arguments, we show that the mini-batch method steadily reduces a natural measure of errors towards zero at the rate of $O(1/K+1/M)$, where $K$ is the number of iterations and $M$ is the batch size.
The rapid growth of deep learning (DL) has spurred interest in enhancing log-based anomaly detection. This approach aims to extract meaning from log events (log message templates) and develop advanced DL models for anomaly detection. However, these DL methods face challenges like heavy reliance on training data, labels, and computational resources due to model complexity. In contrast, traditional machine learning and data mining techniques are less data-dependent and more efficient but less effective than DL. To make log-based anomaly detection more practical, the goal is to enhance traditional techniques to match DL's effectiveness. Previous research in a different domain (linking questions on Stack Overflow) suggests that optimized traditional techniques can rival state-of-the-art DL methods. Drawing inspiration from this concept, we conducted an empirical study. We optimized the unsupervised PCA (Principal Component Analysis), a traditional technique, by incorporating lightweight semantic-based log representation. This addresses the issue of unseen log events in training data, enhancing log representation. Our study compared seven log-based anomaly detection methods, including four DL-based, two traditional, and the optimized PCA technique, using public and industrial datasets. Results indicate that the optimized unsupervised PCA technique achieves similar effectiveness to advanced supervised/semi-supervised DL methods while being more stable with limited training data and resource-efficient. This demonstrates the adaptability and strength of traditional techniques through small yet impactful adaptations.
Dysarthric speech reconstruction (DSR) aims to transform dysarthric speech into normal speech by improving the intelligibility and naturalness. This is a challenging task especially for patients with severe dysarthria and speaking in complex, noisy acoustic environments. To address these challenges, we propose a novel multi-modal framework to utilize visual information, e.g., lip movements, in DSR as extra clues for reconstructing the highly abnormal pronunciations. The multi-modal framework consists of: (i) a multi-modal encoder to extract robust phoneme embeddings from dysarthric speech with auxiliary visual features; (ii) a variance adaptor to infer the normal phoneme duration and pitch contour from the extracted phoneme embeddings; (iii) a speaker encoder to encode the speaker's voice characteristics; and (iv) a mel-decoder to generate the reconstructed mel-spectrogram based on the extracted phoneme embeddings, prosodic features and speaker embeddings. Both objective and subjective evaluations conducted on the commonly used UASpeech corpus show that our proposed approach can achieve significant improvements over baseline systems in terms of speech intelligibility and naturalness, especially for the speakers with more severe symptoms. Compared with original dysarthric speech, the reconstructed speech achieves 42.1\% absolute word error rate reduction for patients with more severe dysarthria levels.
Design technology co-optimization (DTCO) plays a critical role in achieving optimal power, performance, and area (PPA) for advanced semiconductor process development. Cell library characterization is essential in DTCO flow, but traditional methods are time-consuming and costly. To overcome these challenges, we propose a graph neural network (GNN)-based machine learning model for rapid and accurate cell library characterization. Our model incorporates cell structures and demonstrates high prediction accuracy across various process-voltage-temperature (PVT) corners and technology parameters. Validation with 512 unseen technology corners and over one million test data points shows accurate predictions of delay, power, and input pin capacitance for 33 types of cells, with a mean absolute percentage error (MAPE) $\le$ 0.95% and a speed-up of 100X compared with SPICE simulations. Additionally, we investigate system-level metrics such as worst negative slack (WNS), leakage power, and dynamic power using predictions obtained from the GNN-based model on unseen corners. Our model achieves precise predictions, with absolute error $\le$3.0 ps for WNS, percentage errors $\le$0.60% for leakage power, and $\le$0.99% for dynamic power, when compared to golden reference. With the developed model, we further proposed a fine-grained drive strength interpolation methodology to enhance PPA for small-to-medium-scale designs, resulting in an approximate 1-3% improvement.
The multi-agent reinforcement learning systems (MARL) based on the Markov decision process (MDP) have emerged in many critical applications. To improve the robustness/defense of MARL systems against adversarial attacks, the study of various adversarial attacks on reinforcement learning systems is very important. Previous works on adversarial attacks considered some possible features to attack in MDP, such as the action poisoning attacks, the reward poisoning attacks, and the state perception attacks. In this paper, we propose a brand-new form of attack called the camouflage attack in the MARL systems. In the camouflage attack, the attackers change the appearances of some objects without changing the actual objects themselves; and the camouflaged appearances may look the same to all the targeted recipient (victim) agents. The camouflaged appearances can mislead the recipient agents to misguided actions. We design algorithms that give the optimal camouflage attacks minimizing the rewards of recipient agents. Our numerical and theoretical results show that camouflage attacks can rival the more conventional, but likely more difficult state perception attacks. We also investigate cost-constrained camouflage attacks and showed numerically how cost budgets affect the attack performance.
Promoting behavioural diversity is critical for solving games with non-transitive dynamics where strategic cycles exist, and there is no consistent winner (e.g., Rock-Paper-Scissors). Yet, there is a lack of rigorous treatment for defining diversity and constructing diversity-aware learning dynamics. In this work, we offer a geometric interpretation of behavioural diversity in games and introduce a novel diversity metric based on \emph{determinantal point processes} (DPP). By incorporating the diversity metric into best-response dynamics, we develop \emph{diverse fictitious play} and \emph{diverse policy-space response oracle} for solving normal-form games and open-ended games. We prove the uniqueness of the diverse best response and the convergence of our algorithms on two-player games. Importantly, we show that maximising the DPP-based diversity metric guarantees to enlarge the \emph{gamescape} -- convex polytopes spanned by agents' mixtures of strategies. To validate our diversity-aware solvers, we test on tens of games that show strong non-transitivity. Results suggest that our methods achieve much lower exploitability than state-of-the-art solvers by finding effective and diverse strategies.
Learning with limited data is a key challenge for visual recognition. Few-shot learning methods address this challenge by learning an instance embedding function from seen classes and apply the function to instances from unseen classes with limited labels. This style of transfer learning is task-agnostic: the embedding function is not learned optimally discriminative with respect to the unseen classes, where discerning among them is the target task. In this paper, we propose a novel approach to adapt the embedding model to the target classification task, yielding embeddings that are task-specific and are discriminative. To this end, we employ a type of self-attention mechanism called Transformer to transform the embeddings from task-agnostic to task-specific by focusing on relating instances from the test instances to the training instances in both seen and unseen classes. Our approach also extends to both transductive and generalized few-shot classification, two important settings that have essential use cases. We verify the effectiveness of our model on two standard benchmark few-shot classification datasets --- MiniImageNet and CUB, where our approach demonstrates state-of-the-art empirical performance.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191