Mingtian Zhang

Liyuan Wang, Xingxing Zhang, Qian Li et al. · microsoft-research

Continual learning aims to empower artificial intelligence (AI) with strong adaptability to the real world. For this purpose, a desirable solution should properly balance memory stability with learning plasticity, and acquire sufficient compatibility to capture the observed distributions. Existing advances mainly focus on preserving memory stability to overcome catastrophic forgetting, but remain difficult to flexibly accommodate incremental changes as biological intelligence (BI) does. By modeling a robust Drosophila learning system that actively regulates forgetting with multiple learning modules, here we propose a generic approach that appropriately attenuates old memories in parameter distributions to improve learning plasticity, and accordingly coordinates a multi-learner architecture to ensure solution compatibility. Through extensive theoretical and empirical validation, our approach not only clearly enhances the performance of continual learning, especially over synaptic regularization methods in task-incremental settings, but also potentially advances the understanding of neurological adaptive mechanisms, serving as a novel paradigm to progress AI and BI together.

Harry Jake Cunningham, Giorgio Giannone, Mingtian Zhang et al. · mit

Global convolutions have shown increasing promise as powerful general-purpose sequence models. However, training long convolutions is challenging, and kernel parameterizations must be able to learn long-range dependencies without overfitting. This work introduces reparameterized multi-resolution convolutions ($\texttt{MRConv}$), a novel approach to parameterizing global convolutional kernels for long-sequence modelling. By leveraging multi-resolution convolutions, incorporating structural reparameterization and introducing learnable kernel decay, $\texttt{MRConv}$ learns expressive long-range kernels that perform well across various data modalities. Our experiments demonstrate state-of-the-art performance on the Long Range Arena, Sequential CIFAR, and Speech Commands tasks among convolution models and linear-time transformers. Moreover, we report improved performance on ImageNet classification by replacing 2D convolutions with 1D $\texttt{MRConv}$ layers.

Mingtian Zhang, Peter Hayes, David Barber

The ability of likelihood-based probabilistic models to generalize to unseen data is central to many machine learning applications such as lossless compression. In this work, we study the generalization of a popular class of probabilistic model - the Variational Auto-Encoder (VAE). We discuss the two generalization gaps that affect VAEs and show that overfitting is usually dominated by amortized inference. Based on this observation, we propose a new training objective that improves the generalization of amortized inference. We demonstrate how our method can improve performance in the context of image modeling and lossless compression.

Mingtian Zhang, Oscar Key, Peter Hayes et al.

Score-based divergences have been widely used in machine learning and statistics applications. Despite their empirical success, a blindness problem has been observed when using these for multi-modal distributions. In this work, we discuss the blindness problem and propose a new family of divergences that can mitigate the blindness problem. We illustrate our proposed divergence in the context of density estimation and report improved performance compared to traditional approaches.

Mingtian Zhang, Tim Z. Xiao, Brooks Paige et al.

Latent variable models like the Variational Auto-Encoder (VAE) are commonly used to learn representations of images. However, for downstream tasks like semantic classification, the representations learned by VAE are less competitive than other non-latent variable models. This has led to some speculations that latent variable models may be fundamentally unsuitable for representation learning. In this work, we study what properties are required for good representations and how different VAE structure choices could affect the learned properties. We show that by using a decoder that prefers to learn local features, the remaining global features can be well captured by the latent, which significantly improves performance of a downstream classification task. We further apply the proposed model to semi-supervised learning tasks and demonstrate improvements in data efficiency.

Mingtian Zhang, Andi Zhang, Tim Z. Xiao et al.

Density-based Out-of-distribution (OOD) detection has recently been shown unreliable for the task of detecting OOD images. Various density ratio based approaches achieve good empirical performance, however methods typically lack a principled probabilistic modelling explanation. In this work, we propose to unify density ratio based methods under a novel framework that builds energy-based models and employs differing base distributions. Under our framework, the density ratio can be viewed as the unnormalized density of an implicit semantic distribution. Further, we propose to directly estimate the density ratio of a data sample through class ratio estimation. We report competitive results on OOD image problems in comparison with recent work that alternatively requires training of deep generative models for the task. Our approach enables a simple and yet effective path towards solving the OOD detection problem.

Madhav Agarwal, Mingtian Zhang, Laura Sevilla-Lara et al.

Speech-driven talking heads have recently emerged and enable interactive avatars. However, real-world applications are limited, as current methods achieve high visual fidelity but slow or fast yet temporally unstable. Diffusion methods provide realistic image generation, yet struggle with oneshot settings. Gaussian Splatting approaches are real-time, yet inaccuracies in facial tracking, or inconsistent Gaussian mappings, lead to unstable outputs and video artifacts that are detrimental to realistic use cases. We address this problem by mapping Gaussian Splatting using 3D Morphable Models to generate person-specific avatars. We introduce transformer-based prediction of model parameters, directly from audio, to drive temporal consistency. From monocular video and independent audio speech inputs, our method enables generation of real-time talking head videos where we report competitive quantitative and qualitative performance.

Peter Hayes, Mingtian Zhang, Raza Habib et al.

We introduce Integrated Weak Learning, a principled framework that integrates weak supervision into the training process of machine learning models. Our approach jointly trains the end-model and a label model that aggregates multiple sources of weak supervision. We introduce a label model that can learn to aggregate weak supervision sources differently for different datapoints and takes into consideration the performance of the end-model during training. We show that our approach outperforms existing weak learning techniques across a set of 6 benchmark classification datasets. When both a small amount of labeled data and weak supervision are present the increase in performance is both consistent and large, reliably getting a 2-5 point test F1 score gain over non-integrated methods.

Andi Zhang, Mingtian Zhang, Damon Wischik

We propose a probabilistic perspective on adversarial examples, allowing us to embed subjective understanding of semantics as a distribution into the process of generating adversarial examples, in a principled manner. Despite significant pixel-level modifications compared to traditional adversarial attacks, our method preserves the overall semantics of the image, making the changes difficult for humans to detect. This extensive pixel-level modification enhances our method's ability to deceive classifiers designed to defend against adversarial attacks. Our empirical findings indicate that the proposed methods achieve higher success rates in circumventing adversarial defense mechanisms, while remaining difficult for human observers to detect.

William Muldrew, Peter Hayes, Mingtian Zhang et al.

As large language models (LLMs) become more capable, fine-tuning techniques for aligning with human intent are increasingly important. A key consideration for aligning these models is how to most effectively use human resources, or model resources in the case where LLMs themselves are used as oracles. Reinforcement learning from Human or AI preferences (RLHF/RLAIF) is the most prominent example of such a technique, but is complex and often unstable. Direct Preference Optimization (DPO) has recently been proposed as a simpler and more stable alternative. In this work, we develop an active learning strategy for DPO to make better use of preference labels. We propose a practical acquisition function for prompt/completion pairs based on the predictive entropy of the language model and a measure of certainty of the implicit preference model optimized by DPO. We demonstrate how our approach improves both the rate of learning and final performance of fine-tuning on pairwise preference data.

Wenlin Chen, Mingtian Zhang, Brooks Paige et al. · cambridge

The inadequate mixing of conventional Markov Chain Monte Carlo (MCMC) methods for multi-modal distributions presents a significant challenge in practical applications such as Bayesian inference and molecular dynamics. Addressing this, we propose Diffusive Gibbs Sampling (DiGS), an innovative family of sampling methods designed for effective sampling from distributions characterized by distant and disconnected modes. DiGS integrates recent developments in diffusion models, leveraging Gaussian convolution to create an auxiliary noisy distribution that bridges isolated modes in the original space and applying Gibbs sampling to alternately draw samples from both spaces. A novel Metropolis-within-Gibbs scheme is proposed to enhance mixing in the denoising sampling step. DiGS exhibits a better mixing property for sampling multi-modal distributions than state-of-the-art methods such as parallel tempering, attaining substantially improved performance across various tasks, including mixtures of Gaussians, Bayesian neural networks and molecular dynamics.

Jiajun He, Wenlin Chen, Mingtian Zhang et al. · cambridge

Training generative models to sample from unnormalized density functions is an important and challenging task in machine learning. Traditional training methods often rely on the reverse Kullback-Leibler (KL) divergence due to its tractability. However, the mode-seeking behavior of reverse KL hinders effective approximation of multi-modal target distributions. To address this, we propose to minimize the reverse KL along diffusion trajectories of both model and target densities. We refer to this objective as the reverse diffusive KL divergence, which allows the model to capture multiple modes. Leveraging this objective, we train neural samplers that can efficiently generate samples from the target distribution in one step. We demonstrate that our method enhances sampling performance across various Boltzmann distributions, including both synthetic multi-modal densities and n-body particle systems.

Mingtian Zhang, Wenlin Chen, Jiajun He et al. · cambridge

Recent advances in training one-step diffusion models typically follow a two-stage pipeline: first training a teacher diffusion model and then distilling it into a one-step student model. This process often depends on both the teacher's score function for supervision and its weights for initializing the student model. In this paper, we explore whether one-step diffusion models can be trained directly without this distillation procedure. We introduce a family of new training methods that entirely forgo teacher score supervision, yet outperforms most teacher-guided distillation approaches. This suggests that score supervision is not essential for effective training of one-step diffusion models. However, we find that initializing the student model with the teacher's weights remains critical. Surprisingly, the key advantage of teacher initialization is not due to better latent-to-output mappings, but rather the rich set of feature representations across different noise levels that the teacher diffusion model provides. These insights take us one step closer towards training one-step diffusion models without distillation and provide a better understanding of the roles of teacher supervision and initialization in the distillation process.

Mingtian Zhang, Shawn Lan, Peter Hayes et al.

Retrieval Augmented Generation (RAG) has emerged as an effective solution for mitigating hallucinations in Large Language Models (LLMs). The retrieval stage in RAG typically involves a pre-trained embedding model, which converts queries and passages into vectors to capture their semantics. However, a standard pre-trained embedding model may exhibit sub-optimal performance when applied to specific domain knowledge, necessitating fine-tuning. This paper addresses scenarios where the embeddings are only available from a black-box model. We introduce Model augmented fine-tuning (Mafin) -- a novel approach for fine-tuning a black-box embedding model by augmenting it with a trainable embedding model. Our results demonstrate that Mafin significantly enhances the performance of the black-box embeddings by only requiring the training of a small augmented model. We validate the effectiveness of our method on both labeled and unlabeled datasets, illustrating its broad applicability and efficiency.

Leyang Wang, Mingtian Zhang, Zijing Ou et al.

Recently, diffusion distillation methods have compressed thousand-step teacher diffusion models into one-step student generators while preserving sample quality. Most existing approaches train the student model using a diffusive divergence whose gradient is approximated via the student's score function, learned through denoising score matching (DSM). Since DSM training is imperfect, the resulting gradient estimate is inevitably biased, leading to sub-optimal performance. In this paper, we propose VarDiU (pronounced /va:rdju:/), a Variational Diffusive Upper Bound that admits an unbiased gradient estimator and can be directly applied to diffusion distillation. Using this objective, we compare our method with Diff-Instruct and demonstrate that it achieves higher generation quality and enables a more efficient and stable training procedure for one-step diffusion distillation.

Zijing Ou, Mingtian Zhang, Andi Zhang et al.

The probabilistic diffusion model has become highly effective across various domains. Typically, sampling from a diffusion model involves using a denoising distribution characterized by a Gaussian with a learned mean and either fixed or learned covariances. In this paper, we leverage the recently proposed covariance moment matching technique and introduce a novel method for learning the diagonal covariance. Unlike traditional data-driven diagonal covariance approximation approaches, our method involves directly regressing the optimal diagonal analytic covariance using a new, unbiased objective named Optimal Covariance Matching (OCM). This approach can significantly reduce the approximation error in covariance prediction. We demonstrate how our method can substantially enhance the sampling efficiency, recall rate and likelihood of commonly used diffusion models.

Mingtian Zhang, Alex Hawkins-Hooker, Brooks Paige et al.

Energy-Based Models (EBMs) offer a versatile framework for modeling complex data distributions. However, training and sampling from EBMs continue to pose significant challenges. The widely-used Denoising Score Matching (DSM) method for scalable EBM training suffers from inconsistency issues, causing the energy model to learn a `noisy' data distribution. In this work, we propose an efficient sampling framework: (pseudo)-Gibbs sampling with moment matching, which enables effective sampling from the underlying clean model when given a `noisy' model that has been well-trained via DSM. We explore the benefits of our approach compared to related methods and demonstrate how to scale the method to high-dimensional datasets.

Mingtian Zhang, James Townsend, Ning Kang et al.

The recently proposed Neural Local Lossless Compression (NeLLoC), which is based on a local autoregressive model, has achieved state-of-the-art (SOTA) out-of-distribution (OOD) generalization performance in the image compression task. In addition to the encouragement of OOD generalization, the local model also allows parallel inference in the decoding stage. In this paper, we propose two parallelization schemes for local autoregressive models. We discuss the practicalities of implementing the schemes and provide experimental evidence of significant gains in compression runtime compared to the previous, non-parallel implementation.

Liyuan Wang, Mingtian Zhang, Zhongfan Jia et al.

Continual learning aims to learn a sequence of tasks from dynamic data distributions. Without accessing to the old training samples, knowledge transfer from the old tasks to each new task is difficult to determine, which might be either positive or negative. If the old knowledge interferes with the learning of a new task, i.e., the forward knowledge transfer is negative, then precisely remembering the old tasks will further aggravate the interference, thus decreasing the performance of continual learning. By contrast, biological neural networks can actively forget the old knowledge that conflicts with the learning of a new experience, through regulating the learning-triggered synaptic expansion and synaptic convergence. Inspired by the biological active forgetting, we propose to actively forget the old knowledge that limits the learning of new tasks to benefit continual learning. Under the framework of Bayesian continual learning, we develop a novel approach named Active Forgetting with synaptic Expansion-Convergence (AFEC). Our method dynamically expands parameters to learn each new task and then selectively combines them, which is formally consistent with the underlying mechanism of biological active forgetting. We extensively evaluate AFEC on a variety of continual learning benchmarks, including CIFAR-10 regression tasks, visual classification tasks and Atari reinforcement tasks, where AFEC effectively improves the learning of new tasks and achieves the state-of-the-art performance in a plug-and-play way.

Mingtian Zhang, Yitong Sun, Chen Zhang et al.

Flow-based models typically define a latent space with dimensionality identical to the observational space. In many problems, however, the data does not populate the full ambient data space that they natively reside in, rather inhabiting a lower-dimensional manifold. In such scenarios, flow-based models are unable to represent data structures exactly as their densities will always have support off the data manifold, potentially resulting in degradation of model performance. To address this issue, we propose to learn a manifold prior for flow models that leverage the recently proposed spread divergence towards fixing the crucial problem; the KL divergence and maximum likelihood estimation are ill-defined for manifold learning. In addition to improving both sample quality and representation quality, an auxiliary benefit enabled by our approach is the ability to identify the intrinsic dimension of the manifold distribution.

Mingtian Zhang, Andi Zhang, Steven McDonagh

Out-of-distribution (OOD) detection and lossless compression constitute two problems that can be solved by the training of probabilistic models on a first dataset with subsequent likelihood evaluation on a second dataset, where data distributions differ. By defining the generalization of probabilistic models in terms of likelihood we show that, in the case of image models, the OOD generalization ability is dominated by local features. This motivates our proposal of a Local Autoregressive model that exclusively models local image features towards improving OOD performance. We apply the proposed model to OOD detection tasks and achieve state-of-the-art unsupervised OOD detection performance without the introduction of additional data. Additionally, we employ our model to build a new lossless image compressor: NeLLoC (Neural Local Lossless Compressor) and report state-of-the-art compression rates and model size.

Mohammed Amin Abdullah, Hang Ren, Haitham Bou Ammar et al.

Reinforcement learning algorithms, though successful, tend to over-fit to training environments hampering their application to the real-world. This paper proposes $\text{W}\text{R}^{2}\text{L}$ -- a robust reinforcement learning algorithm with significant robust performance on low and high-dimensional control tasks. Our method formalises robust reinforcement learning as a novel min-max game with a Wasserstein constraint for a correct and convergent solver. Apart from the formulation, we also propose an efficient and scalable solver following a novel zero-order optimisation method that we believe can be useful to numerical optimisation in general. We empirically demonstrate significant gains compared to standard and robust state-of-the-art algorithms on high-dimensional MuJuCo environments.

Mingtian Zhang, Thomas Bird, Raza Habib et al.

Probabilistic models are often trained by maximum likelihood, which corresponds to minimizing a specific f-divergence between the model and data distribution. In light of recent successes in training Generative Adversarial Networks, alternative non-likelihood training criteria have been proposed. Whilst not necessarily statistically efficient, these alternatives may better match user requirements such as sharp image generation. A general variational method for training probabilistic latent variable models using maximum likelihood is well established; however, how to train latent variable models using other f-divergences is comparatively unknown. We discuss a variational approach that, when combined with the recently introduced Spread Divergence, can be applied to train a large class of latent variable models using any f-divergence.

Mingtian Zhang, Peter Hayes, Tom Bird et al.

For distributions $\mathbb{P}$ and $\mathbb{Q}$ with different supports or undefined densities, the divergence $\textrm{D}(\mathbb{P}||\mathbb{Q})$ may not exist. We define a Spread Divergence $\tilde{\textrm{D}}(\mathbb{P}||\mathbb{Q})$ on modified $\mathbb{P}$ and $\mathbb{Q}$ and describe sufficient conditions for the existence of such a divergence. We demonstrate how to maximize the discriminatory power of a given divergence by parameterizing and learning the spread. We also give examples of using a Spread Divergence to train implicit generative models, including linear models (Independent Components Analysis) and non-linear models (Deep Generative Networks).