With the rapid growth of large language models (LLMs), there is increasing demand for memory and computation in LLMs. Recent efforts on post-training pruning of LLMs aim to reduce the model size and computation requirements, yet the performance is still sub-optimal. In this paper, we present a plug-and-play solution for post-training pruning of LLMs.The proposed solution has two innovative components: 1) **Relative Importance and Activations (RIA)**, a new pruning metric that jointly considers the weight and activations efficiently on LLMs, and 2) **Channel Permutation**, a new approach to maximally preserves important weights under N:M sparsity.The two proposed components can be readily combined to further enhance the N:M semi-structured pruning of LLMs.Our empirical experiments show that RIA alone can already surpass all existing post-training pruning methods on prevalent LLMs, e.g., LLaMA ranging from 7B to 65B. Furthermore, N:M semi-structured pruning with channel permutation can even outperform the original LLaMA2-70B on zero-shot tasks, together with practical speed-up on specific hardware.Our code is available at: https://github.com/biomedical-cybernetics/Relative-importance-and-activation-pruning
论文检索
输入标题、作者或关键词，从 12,113 篇学术成果中精准定位
Recent works have shown that Large Language Models (LLMs) could empower traditional neuro-symbolic models via programming capabilities to translate languages into module descriptions, thus achieving strong visual reasoning results while maintaining the model’s transparency and efficiency. However, these models usually exhaustively generate the entire code snippet given each new instance of a task, which is extremely ineffective. On the contrary, human beings gradually acquire knowledge that can be reused and grow into more profound skills for fast generalization to new tasks since we are an infant. Inspired by this, we propose generative neuro-symbolic visual reasoning by growing and reusing modules. Specifically, our model consists of three unique stages, module initialization, module generation, and module execution. First, given a vision-language task, we adopt LLMs to examine whether we could reuse and grow over established modules to handle this new task. If not, we initialize a new module needed by the task and specify the inputs and outputs of this new module. After that, the new module is created by querying LLMs to generate corresponding code snippets that match the requirements. In order to get a better sense of the new module’s ability, we treat few-shot training examples as test cases to see if our new module could pass these cases. If yes, the new module is added to the module library for future reuse. Finally, we evaluate the performance of our model on the testing set by executing the parsed programs with the newly made visual modules to get the results. We find the proposed GENOME model possesses several advantages. First, it performs competitively on standard tasks like visual question answering and referring expression comprehension; Second, the visual modules learned from one task can be seamlessly transferred to new tasks; Last but not least, it is able to adapt to new visual reasoning tasks by observing a few training examples and reusing modules.
Deep Learning · Sequential Models, Time series
*Not all positive pairs are beneficial to time series contrastive learning*. In this paper, we study two types of bad positive pairs that can impair the quality of time series representation learned through contrastive learning: the noisy positive pair and the faulty positive pair. We observe that, with the presence of noisy positive pairs, the model tends to simply learn the pattern of noise (Noisy Alignment). Meanwhile, when faulty positive pairs arise, the model wastes considerable amount of effort aligning non-representative patterns (Faulty Alignment). To address this problem, we propose a Dynamic Bad Pair Mining (DBPM) algorithm, which reliably identifies and suppresses bad positive pairs in time series contrastive learning. Specifically, DBPM utilizes a memory module to dynamically track the training behavior of each positive pair along training process. This allows us to identify potential bad positive pairs at each epoch based on their historical training behaviors. The identified bad pairs are subsequently down-weighted through a transformation module, thereby mitigating their negative impact on the representation learning process. DBPM is a simple algorithm designed as a lightweight **plug-in** without learnable parameters to enhance the performance of existing state-of-the-art methods. Through extensive experiments conducted on four large-scale, real-world time series datasets, we demonstrate DBPM's efficacy in mitigating the adverse effects of bad positive pairs.
Deep Learning · Other Representation Learning
Recognition problems in long-tailed data, in which the sample size per class is heavily skewed, have gained importance because the distribution of the sample size per class in a dataset is generally exponential unless the sample size is intentionally adjusted. Various methods have been devised to address these problems.Recently, weight balancing, which combines well-known classical regularization techniques with two-stage training, has been proposed. Despite its simplicity, it is known for its high performance compared with existing methods devised in various ways.However, there is a lack of understanding as to why this method is effective for long-tailed data. In this study, we analyze weight balancing by focusing on neural collapse and the cone effect at each training stage and found that it can be decomposed into an increase in Fisher's discriminant ratio of the feature extractor caused by weight decay and cross entropy loss and implicit logit adjustment caused by weight decay and class-balanced loss. Our analysis enables the training method to be further simplified by reducing the number of training stages to one while increasing accuracy. Code is available at https://github.com/HN410/Exploring-Weight-Balancing-on-Long-Tailed-Recognition-Problem.
Deep Learning · Other Representation Learning
Similarity-based representation learning has shown impressive capabilities in both supervised (e.g., metric learning) and unsupervised (e.g., contrastive learning) scenarios. Existing approaches effectively constrained the representation difference (i.e., the disagreement between the embeddings of two instances) to fit the corresponding (pseudo) similarity supervision. However, most of them can hardly restrict the variation of representation difference, sometimes leading to overfitting results where the clusters are disordered by drastically changed differences. In this paper, we thus propose a novel difference alignment regularization (DAR) to encourage all representation differences between inter-class instances to be as close as possible, so that the learning algorithm can produce consistent differences to distinguish data points from each other. To this end, we construct a new cross-total-variation (CTV) norm to measure the divergence among representation differences, and we convert it into an equivalent stochastic form for easy optimization. Then, we integrate the proposed regularizer into the empirical loss for difference-aligned similarity learning (DASL), shrinking the hypothesis space and alleviating overfitting. Theoretically, we prove that our regularizer tightens the error bound of the traditional similarity learning. Experiments on multi-domain data demonstrate the superiority of DASL over existing approaches in both supervised metric learning and unsupervised contrastive learning tasks.
Applications · Computer Vision
More distinguishable and consistent pixel features for each category will benefit the semantic segmentation under various settings.Existing efforts to mine better pixel-level features attempt to explicitly model the categorical distribution, which fails to achieve optimal due to the significant pixel feature variance.Moreover, prior research endeavors have scarcely delved into the thorough analysis and meticulous handling of pixel-level variance, leaving semantic segmentation at a coarse granularity.In this work, we analyze the causes of pixel-level variance and introduce the concept of $\textbf{pixel learning}$ to concentrate on the tailored learning process of pixels to handle the pixel-level variance, enhancing the per-pixel recognition capability of segmentation models.Under the context of the pixel learning scheme, each image is viewed as a distribution of pixels, and pixel learning aims to pursue consistent pixel representation inside an image, continuously align pixels from different images (distributions), and eventually achieve consistent pixel representation for each category, even cross-domains.We proposed a pure pixel-level learning framework, namely PiXL, which consists of a pixel partition module to divide pixels into sub-domains, a prototype generation, a selection module to prepare targets for subsequent alignment, and a pixel alignment module to guarantee pixel feature consistency intra-/inter-images, and inter-domains.Extensive evaluations of multiple learning paradigms, including unsupervised domain adaptation and semi-/fully-supervised segmentation, show that PiXL outperforms state-of-the-art performances, especially when annotated images are scarce.Visualization of the embedding space further demonstrates that pixel learning attains a superior representation of pixel features.The code is available at https://github.com/ChenGan-JS/PiXL.
Applications · Neuroscience, Cognitive Science
Motivated by a recent neuroscientific hypothesis, some theoretical studies have accounted for neural cognitive maps in the rodent hippocampal formation as a representation of the general relational structure across task environments. However, despite their remarkable results, it is unclear whether their account can be extended to more general settings beyond spatial random-walk tasks in 2D environments. To address this question, we construct a novel cognitive model that performs memory-based relational decision-making tasks, inspired by previous human studies, for learning abstract structures in non-spatial relations. Building on previous approaches of modular architecture, we develop a learning algorithm that performs reward-guided search for representation of abstract relations, while dynamically maintaining their binding to concrete entities using our specific memory mechanism enabling content replacement. Our experiments show (i) the capability of our model to capture relational structures that can generalize over new domains with unseen entities, (ii) the difficulty of our task that leads previous models, including Neural Turing Machine and vanilla Transformer, to complete failure, and (iii) the similarity of performance and internal representations of our model to recent human behavioral and fMRI experimental data in the human hippocampal formation.
Deep Learning · Other Representation Learning
We present ReMasker, a new method of imputing missing values in tabular data by extending the masked autoencoding framework. Compared with prior work, ReMasker is extremely simple -- besides the missing values (i.e., naturally masked), we randomly "re-mask" another set of values, optimize the autoencoder by reconstructing this re-masked set, and apply the trained model to predict the missing values; and yet highly effective -- with extensive evaluation on benchmark datasets, we show that ReMasker performs on par with or outperforms state-of-the-art methods in terms of both imputation fidelity and utility under various missingness settings, while its performance advantage often increases with the ratio of missing data. We further explore theoretical justification for its effectiveness, showing that ReMasker tends to learn missingness-invariant representations of tabular data. Our findings indicate that masked modeling represents a promising direction for further research on tabular data imputation. The code is publicly available.
Deep Learning · Large Language Models
Quantization is an indispensable technique for serving Large Language Models (LLMs) and has recently found its way into LoRA fine-tuning (Dettmers et al., 2023). In this work we focus on the scenario where quantization and LoRA fine- tuning are applied together on a pre-trained model. In such cases it is common to observe a consistent gap in the performance on downstream tasks between full fine-tuning and quantization plus LoRA fine-tuning approach. In response, we propose LoftQ (LoRA-Fine-Tuning-aware Quantization), a novel quantization framework that simultaneously quantizes an LLM and finds a proper low-rank initialization for LoRA fine-tuning. Such an initialization alleviates the discrep- ancy between the quantized and full-precision model and significantly improves the generalization in downstream tasks. We evaluate our method on natural lan- guage understanding, question answering, summarization, and natural language generation tasks. Experiments show that our method is highly effective and out- performs existing quantization methods, especially in the challenging 2-bit and 2/4-bit mixed precision regimes. We will release our code.
Deep Learning · Generative Models and Autoencoders
Learning disentangled representations for time series is a promising path to facilitate reliable generalization to in- and out-of distribution (OOD), offering benefits like feature derivation and improved interpretability and fairness, thereby enhancing downstream tasks. We focus on disentangled representation learning for home appliance electricity usage, enabling users to understand and optimize their consumption for a reduced carbon footprint. Our approach frames the problem as disentangling each attribute's role in total consumption. Unlike existing methods assuming attribute independence which leads to non-identiability, we acknowledge real-world time series attribute correlations, learned up to a smooth bijection using contrastive learning and a single autoencoder. To address this, we propose a Disentanglement under Independence-Of-Support via Contrastive Learning (DIOSC), facilitating representation generalization across diverse correlated scenarios. Our method utilizes innovative \textit{l}-variational inference layers with self-attention, effectively addressing temporal dependencies across bottom-up and top-down networks. We find that DIOSC can enhance the task of representation of time series electricity consumption. We introduce TDS (Time Disentangling Score) to gauge disentanglement quality. TDS reliably reflects disentanglement performance, making it a valuable metric for evaluating time series representations disentanglement. Code available at https://institut-polytechnique-de-paris.github.io/time-disentanglement-lib.
We investigate learning the equilibria in non-stationary multi-agent systems and address the challenges that differentiate multi-agent learning from single-agent learning. Specifically, we focus on games with bandit feedback, where testing an equilibrium can result in substantial regret even when the gap to be tested is small, and the existence of multiple optimal solutions (equilibria) in stationary games poses extra challenges. To overcome these obstacles, we propose a versatile black-box approach applicable to a broad spectrum of problems, such as general-sum games, potential games, and Markov games, when equipped with appropriate learning and testing oracles for stationary environments. Our algorithms can achieve $\widetilde{O}\left(\Delta^{1/4}T^{3/4}\right)$ regret when the degree of nonstationarity, as measured by total variation $\Delta$, is known, and $\widetilde{O}\left(\Delta^{1/5}T^{4/5}\right)$ regret when $\Delta$ is unknown, where $T$ is the number of rounds. Meanwhile, our algorithm inherits the favorable dependence on number of agents from the oracles. As a side contribution that may be independent of interest, we show how to test for various types of equilibria by a black-box reduction to single-agent learning, which includes Nash equilibria, correlated equilibria, and coarse correlated equilibria.
We present a new technique to enhance the robustness of imitation learning methods by generating corrective data to account for compounding error and disturbances. While existing methods rely on interactive expert labeling, additional offline datasets, or domain-specific invariances, our approach requires minimal additional assumptions beyond expert data. The key insight is to leverage local continuity in the environment dynamics. Our method first constructs a dynamics model from the expert demonstration, enforcing local Lipschitz continuity while skipping the discontinuous regions. In the locally continuous regions, this model allows us to generate corrective labels within the neighborhood of the demonstrations but beyond the actual set of states and actions in the dataset. Training on this augmented data enhances the agent's ability to recover from perturbations and deal with compounding error. We demonstrate the effectiveness of our generated labels through experiments in a variety of robotics domains that have distinct forms of continuity and discontinuity, including classic control, drone flying, high-dimensional navigation, locomotion, and tabletop manipulation.
Theory · Online Learning and Bandits
Cascading bandits have gained popularity in recent years due to their applicability to recommendation systems and online advertising. In the cascading bandit model, at each timestep, an agent recommends an ordered subset of items (called an item list) from a pool of items, each associated with an unknown attraction probability. Then, the user examines the list, and clicks the first attractive item (if any), and after that, the agent receives a reward. The goal of the agent is to maximize the expected cumulative reward. However, the prior literature on cascading bandits ignores the influences of user states (e.g., historical behaviors) on recommendations and the change of states as the session proceeds. Motivated by this fact, we propose a generalized cascading RL framework, which considers the impact of user states and state transition into decisions. In cascading RL, we need to select items not only with large attraction probabilities but also leading to good successor states. This imposes a huge computational challenge due to the combinatorial action space. To tackle this challenge, we delve into the properties of value functions, and design an oracle BestPerm to efficiently find the optimal item list. Equipped with BestPerm, we develop two algorithms CascadingVI and CascadingBPI, which are both computationally-efficient and sample-efficient, and provide near-optimal regret and sample complexity guarantees. Furthermore, we present experiments to show the improved computational and sample efficiencies of our algorithms compared to straightforward adaptations of existing RL algorithms in practice.
In light of the widespread success of generative models, a significant amount of research has gone into speeding up their sampling time. However, generative models are often sampled multiple times to obtain a diverse set incurring a cost that is orthogonal to sampling time. We tackle the question of how to improve diversity and sample efficiency by moving beyond the common assumption of independent samples. We propose particle guidance, an extension of diffusion-based generative sampling where a joint-particle time-evolving potential enforces diversity. We analyze theoretically the joint distribution that particle guidance generates, how to learn a potential that achieves optimal diversity, and the connections with methods in other disciplines. Empirically, we test the framework both in the setting of conditional image generation, where we are able to increase diversity without affecting quality, and molecular conformer generation, where we reduce the state-of-the-art median error by 13% on average.
Evaluating the performance of a well-trained GNN model on real-world graphs is a pivotal step for reliable GNN online deployment and serving. Due to a lack of test node labels and unknown potential training-test graph data distribution shifts, conventional model evaluation encounters limitations in calculating performance metrics (e.g., test error) and measuring graph data-level discrepancies, particularly when the training graph used for developing GNNs remains unobserved during test time.In this paper, we study a new research problem, online GNN evaluation, which aims to provide valuable insights into the well-trained GNNs's ability to effectively generalize to real-world unlabeled graphs under the test-time graph distribution shifts.Concretely, we develop an effective learning behavior discrepancy score, dubbed LeBeD, to estimate the test-time generalization errors of well-trained GNN models. Through a novel GNN re-training strategy with a parameter-free optimality criterion, the proposed LeBeD comprehensively integrates learning behavior discrepancies from both node prediction and structure reconstruction perspectives.This enables the effective evaluation of the well-trained GNNs' ability to capture test node semantics and structural representations, making it an expressive metric for estimating the generalization error in online GNN evaluation.Extensive experiments on real-world test graphs under diverse graph distribution shifts could verify the effectiveness of the proposed method, revealing its strong correlation with ground-truth test errors on various well-trained GNN models.
Reinforcement Learning · Batch/Offline
In this study, we investigate the DIstribution Correction Estimation (DICE) methods, an important line of work in offline reinforcement learning (RL) and imitation learning (IL). DICE-based methods impose state-action-level behavior constraint, which is an ideal choice for offline learning. However, they typically perform much worse than current state-of-the-art (SOTA) methods that solely use action-level behavior constraint. After revisiting DICE-based methods, we find there exist two gradient terms when learning the value function using true-gradient update: forward gradient (taken on the current state) and backward gradient (taken on the next state). Using forward gradient bears a large similarity to many offline RL methods, and thus can be regarded as applying action-level constraint. However, directly adding the backward gradient may degenerate or cancel out its effect if these two gradients have conflicting directions. To resolve this issue, we propose a simple yet effective modification that projects the backward gradient onto the normal plane of the forward gradient, resulting in an orthogonal-gradient update, a new learning rule for DICE-based methods. We conduct thorough theoretical analyses and find that the projected backward gradient brings state-level behavior regularization, which reveals the mystery of DICE-based methods: the value learning objective does try to impose state-action-level constraint, but needs to be used in a corrected way. Through toy examples and extensive experiments on complex offline RL and IL tasks, we demonstrate that DICE-based methods using orthogonal-gradient updates achieve SOTA performance and great robustness.
We propose Joint MLP/Attention (JoMA) dynamics, a novel mathematical framework to understand the training procedure of multilayer Transformer architectures. This is achieved by integrating out the self-attention layer in Transformers, producing a modified dynamics of MLP layers only. JoMA removes unrealistic assumptions in previous analysis (e.g., lack of residual connection), and predicts that the attention first becomes sparse (to learn salient tokens), then dense (to learn less salient tokens) in the presence of nonlinear activations, while in the linear case, it is consistent with existing works. We leverage JoMA to qualitatively explains how tokens are combined to form hierarchies in multilayer Transformers, when the input tokens are generated by a latent hierarchical generative model. Experiments on models trained from real-world dataset (Wikitext2/Wikitext103) and various pre- trained models (OPT, Pythia) verify our theoretical findings. The code is at https://github.com/facebookresearch/luckmatters/tree/yuandong3.
Bilevel optimization is an important formulation for many machine learning problems, such as meta-learning and hyperparameter optimization. Current bilevel optimization algorithms assume that the gradient of the upper-level function is Lipschitz (i.e., the upper-level function has a bounded smoothness parameter). However, recent studies reveal that certain neural networks such as recurrent neural networks (RNNs) and long-short-term memory networks (LSTMs) exhibit potential unbounded smoothness, rendering conventional bilevel optimization algorithms unsuitable for these neural networks. In this paper, we design a new bilevel optimization algorithm, namely BO-REP, to address this challenge. This algorithm updates the upper-level variable using normalized momentum and incorporates two novel techniques for updating the lower-level variable: \textit{initialization refinement} and \textit{periodic updates}. Specifically, once the upper-level variable is initialized, a subroutine is invoked to obtain a refined estimate of the corresponding optimal lower-level variable, and the lower-level variable is updated only after every specific period instead of each iteration. When the upper-level problem is nonconvex and unbounded smooth, and the lower-level problem is strongly convex, we prove that our algorithm requires $\widetilde{O}(1/\epsilon^4)$ \footnote{Here $\widetilde{O}(\cdot)$ compresses logarithmic factors of $1/\epsilon$ and $1/\delta$, where $\delta\in(0,1)$ denotes the failure probability.} iterations to find an $\epsilon$-stationary point in the stochastic setting, where each iteration involves calling a stochastic gradient or Hessian-vector product oracle. Notably, this result matches the state-of-the-art complexity results under the bounded smoothness setting and without mean-squared smoothness of the stochastic gradient, up to logarithmic factors. Our proof relies on novel technical lemmas for the periodically updated lower-level variable, which are of independent interest. Our experiments on hyper-representation learning, hyperparameter optimization, and data hyper-cleaning for text classification tasks demonstrate the effectiveness of our proposed algorithm. The code is available at [https://github.com/MingruiLiu-ML-Lab/Bilevel-Optimization-under-Unbounded-Smoothness](https://github.com/MingruiLiu-ML-Lab/Bilevel-Optimization-under-Unbounded-Smoothness).
Deep Learning · Robustness
Supervised learning datasets may contain multiple cues that explain the training set equally well, i.e., learning any of them would lead to the correct predictions on the training data. However, many of them can be spurious, i.e., lose their predictive power under a distribution shift and consequently fail to generalize to out-of-distribution (OOD) data. Recently developed "diversification" methods (Lee et al., 2023; Pagliardini et al., 2023) approach this problem by finding multiple diverse hypotheses that rely on different features. This paper aims to study this class of methods and identify the key components contributing to their OOD generalization abilities.We show that (1) diversification methods are highly sensitive to the distribution of the unlabeled data used for diversification and can underperform significantly when away from a method-specific sweet spot. (2) Diversification alone is insufficient for OOD generalization. The choice of the used learning algorithm, e.g., the model's architecture and pretraining, is crucial. In standard experiments (classification on Waterbirds and Office-Home datasets), using the second-best choice leads to an up to 20\% absolute drop in accuracy. (3) The optimal choice of learning algorithm depends on the unlabeled data and vice versa i.e. they are co-dependent. (4) Finally, we show that, in practice, the above pitfalls cannot be alleviated by increasing the number of diverse hypotheses, the major feature of diversification methods.These findings provide a clearer understanding of the critical design factors influencing the OOD generalization abilities of diversification methods. They can guide practitioners in how to use the existing methods best and guide researchers in developing new, better ones.
Reinforcement Learning · Deep RL
We introduce AMAGO, an in-context Reinforcement Learning (RL) agent that uses sequence models to tackle the challenges of generalization, long-term memory, and meta-learning. Recent works have shown that off-policy learning can make in-context RL with recurrent policies viable. Nonetheless, these approaches require extensive tuning and limit scalability by creating key bottlenecks in agents' memory capacity, planning horizon, and model size. AMAGO revisits and redesigns the off-policy in-context approach to successfully train long-sequence Transformers over entire rollouts in parallel with end-to-end RL. Our agent is scalable and applicable to a wide range of problems, and we demonstrate its strong performance empirically in meta-RL and long-term memory domains. AMAGO's focus on sparse rewards and off-policy data also allows in-context learning to extend to goal-conditioned problems with challenging exploration. When combined with a multi-goal hindsight relabeling scheme, AMAGO can solve a previously difficult category of open-world domains, where agents complete many possible instructions in procedurally generated environments.