Gonçalo Mordido

Pranshu Malviya, Gonçalo Mordido, Aristide Baratin et al. · deepmind

Adaptive gradient-based optimizers, notably Adam, have left their mark in training large-scale deep learning models, offering fast convergence and robustness to hyperparameter settings. However, they often struggle with generalization, attributed to their tendency to converge to sharp minima in the loss landscape. To address this, we propose a new memory-augmented version of Adam that encourages exploration towards flatter minima by incorporating a buffer of critical momentum terms during training. This buffer prompts the optimizer to overshoot beyond narrow minima, promoting exploration. Through comprehensive analysis in simple settings, we illustrate the efficacy of our approach in increasing exploration and bias towards flatter minima. We empirically demonstrate that it can improve model performance for image classification on ImageNet and CIFAR10/100, language modelling on Penn Treebank, and online learning tasks on TinyImageNet and 5-dataset. Our code is available at \url{https://github.com/chandar-lab/CMOptimizer}.

Kamran Chitsaz, Quentin Fournier, Gonçalo Mordido et al.

The increasing scale of Transformer models has led to an increase in their pre-training computational requirements. While quantization has proven to be effective after pre-training and during fine-tuning, applying quantization in Transformers during pre-training has remained largely unexplored at scale for language modeling. This study aims to explore the impact of quantization for efficient pre-training of Transformers, with a focus on linear layer components. By systematically applying straightforward linear quantization to weights, activations, gradients, and optimizer states, we assess its effects on model efficiency, stability, and performance during training. By offering a comprehensive recipe of effective quantization strategies to be applied during the pre-training of Transformers, we promote high training efficiency from scratch while retaining language modeling ability. Code is available at https://github.com/chandar-lab/EfficientLLMs.

Gonçalo Mordido, Pranshu Malviya, Aristide Baratin et al.

Sharpness-aware minimization (SAM) methods have gained increasing popularity by formulating the problem of minimizing both loss value and loss sharpness as a minimax objective. In this work, we increase the efficiency of the maximization and minimization parts of SAM's objective to achieve a better loss-sharpness trade-off. By taking inspiration from the Lookahead optimizer, which uses multiple descent steps ahead, we propose Lookbehind, which performs multiple ascent steps behind to enhance the maximization step of SAM and find a worst-case perturbation with higher loss. Then, to mitigate the variance in the descent step arising from the gathered gradients across the multiple ascent steps, we employ linear interpolation to refine the minimization step. Lookbehind leads to a myriad of benefits across a variety of tasks. Particularly, we show increased generalization performance, greater robustness against noisy weights, as well as improved learning and less catastrophic forgetting in lifelong learning settings. Our code is available at https://github.com/chandar-lab/Lookbehind-SAM.

Simon Guiroy, Christopher Pal, Gonçalo Mordido et al.

Meta-Learning algorithms for few-shot learning aim to train neural networks capable of generalizing to novel tasks using only a few examples. Early-stopping is critical for performance, halting model training when it reaches optimal generalization to the new task distribution. Early-stopping mechanisms in Meta-Learning typically rely on measuring the model performance on labeled examples from a meta-validation set drawn from the training (source) dataset. This is problematic in few-shot transfer learning settings, where the meta-test set comes from a different target dataset (OOD) and can potentially have a large distributional shift with the meta-validation set. In this work, we propose Activation Based Early-stopping (ABE), an alternative to using validation-based early-stopping for meta-learning. Specifically, we analyze the evolution, during meta-training, of the neural activations at each hidden layer, on a small set of unlabelled support examples from a single task of the target tasks distribution, as this constitutes a minimal and justifiably accessible information from the target problem. Our experiments show that simple, label agnostic statistics on the activations offer an effective way to estimate how the target generalization evolves over time. At each hidden layer, we characterize the activation distributions, from their first and second order moments, then further summarized along the feature dimensions, resulting in a compact yet intuitive characterization in a four-dimensional space. Detecting when, throughout training time, and at which layer, the target activation trajectory diverges from the activation trajectory of the source data, allows us to perform early-stopping and improve generalization in a large array of few-shot transfer learning settings, across different algorithms, source and target datasets.

Jonathan Kern, Sébastien Henwood, Gonçalo Mordido et al.

Memristors enable the computation of matrix-vector multiplications (MVM) in memory and, therefore, show great potential in highly increasing the energy efficiency of deep neural network (DNN) inference accelerators. However, computations in memristors suffer from hardware non-idealities and are subject to different sources of noise that may negatively impact system performance. In this work, we theoretically analyze the mean squared error of DNNs that use memristor crossbars to compute MVM. We take into account both the quantization noise, due to the necessity of reducing the DNN model size, and the programming noise, stemming from the variability during the programming of the memristance value. Simulations on pre-trained DNN models showcase the accuracy of the analytical prediction. Furthermore the proposed method is almost two order of magnitude faster than Monte-Carlo simulation, thus making it possible to optimize the implementation parameters to achieve minimal error for a given power constraint.

Gonçalo Mordido, Sébastien Henwood, Sarath Chandar et al.

Energy-efficient deep neural network (DNN) accelerators are prone to non-idealities that degrade DNN performance at inference time. To mitigate such degradation, existing methods typically add perturbations to the DNN weights during training to simulate inference on noisy hardware. However, this often requires knowledge about the target hardware and leads to a trade-off between DNN performance and robustness, decreasing the former to increase the latter. In this work, we show that applying sharpness-aware training, by optimizing for both the loss value and loss sharpness, significantly improves robustness to noisy hardware at inference time without relying on any assumptions about the target hardware. In particular, we propose a new adaptive sharpness-aware method that conditions the worst-case perturbation of a given weight not only on its magnitude but also on the range of the weight distribution. This is achieved by performing sharpness-aware minimization scaled by outlier minimization (SAMSON). Our approach outperforms existing sharpness-aware training methods both in terms of model generalization performance in noiseless regimes and robustness in noisy settings, as measured on several architectures and datasets.

Gonçalo Mordido, Matthijs Van Keirsbilck, Alexander Keller

We demonstrate that 1x1-convolutions in 1D time-channel separable convolutions may be replaced by constant, sparse random ternary matrices with weights in $\{-1,0,+1\}$. Such layers do not perform any multiplications and do not require training. Moreover, the matrices may be generated on the chip during computation and therefore do not require any memory access. With the same parameter budget, we can afford deeper and more expressive models, improving the Pareto frontiers of existing models on several tasks. For command recognition on Google Speech Commands v1, we improve the state-of-the-art accuracy from $97.21\%$ to $97.41\%$ at the same network size. Alternatively, we can lower the cost of existing models. For speech recognition on Librispeech, we half the number of weights to be trained while only sacrificing about $1\%$ of the floating-point baseline's word error rate.

Gonçalo Mordido, Haojin Yang, Christoph Meinel

The analysis of the compression effects in generative adversarial networks (GANs) after training, i.e. without any fine-tuning, remains an unstudied, albeit important, topic with the increasing trend of their computation and memory requirements. While existing works discuss the difficulty of compressing GANs during training, requiring novel methods designed with the instability of GANs training in mind, we show that existing compression methods (namely clipping and quantization) may be directly applied to compress GANs post-training, without any additional changes. High compression levels may distort the generated set, likely leading to an increase of outliers that may negatively affect the overall assessment of existing k-nearest neighbor (KNN) based metrics. We propose two new precision and recall metrics based on locality-sensitive hashing (LSH), which, on top of increasing the outlier robustness, decrease the complexity of assessing an evaluation sample against $n$ reference samples from $O(n)$ to $O(\log(n))$, if using LSH and KNN, and to $O(1)$, if only applying LSH. We show that low-bit compression of several pre-trained GANs on multiple datasets induces a trade-off between precision and recall, retaining sample quality while sacrificing sample diversity.

Gonçalo Mordido, Christoph Meinel

We propose a family of metrics to assess language generation derived from population estimation methods widely used in ecology. More specifically, we use mark-recapture and maximum-likelihood methods that have been applied over the past several decades to estimate the size of closed populations in the wild. We propose three novel metrics: ME$_\text{Petersen}$ and ME$_\text{CAPTURE}$, which retrieve a single-valued assessment, and ME$_\text{Schnabel}$ which returns a double-valued metric to assess the evaluation set in terms of quality and diversity, separately. In synthetic experiments, our family of methods is sensitive to drops in quality and diversity. Moreover, our methods show a higher correlation to human evaluation than existing metrics on several challenging tasks, namely unconditional language generation, machine translation, and text summarization.

Julian Niedermeier, Gonçalo Mordido, Christoph Meinel

Objective and interpretable metrics to evaluate current artificial intelligent systems are of great importance, not only to analyze the current state of such systems but also to objectively measure progress in the future. In this work, we focus on the evaluation of image generation tasks. We propose a novel approach, called Fuzzy Topology Impact (FTI), that determines both the quality and diversity of an image set using topology representations combined with fuzzy logic. When compared to current evaluation methods, FTI shows better and more stable performance on multiple experiments evaluating the sensitivity to noise, mode dropping and mode inventing.

Gonçalo Mordido, Haojin Yang, Christoph Meinel

We propose to tackle the mode collapse problem in generative adversarial networks (GANs) by using multiple discriminators and assigning a different portion of each minibatch, called microbatch, to each discriminator. We gradually change each discriminator's task from distinguishing between real and fake samples to discriminating samples coming from inside or outside its assigned microbatch by using a diversity parameter $α$. The generator is then forced to promote variety in each minibatch to make the microbatch discrimination harder to achieve by each discriminator. Thus, all models in our framework benefit from having variety in the generated set to reduce their respective losses. We show evidence that our solution promotes sample diversity since early training stages on multiple datasets.

Gonçalo Mordido, Matthijs Van Keirsbilck, Alexander Keller

Low bit-width integer weights and activations are very important for efficient inference, especially with respect to lower power consumption. We propose Monte Carlo methods to quantize the weights and activations of pre-trained neural networks without any re-training. By performing importance sampling we obtain quantized low bit-width integer values from full-precision weights and activations. The precision, sparsity, and complexity are easily configurable by the amount of sampling performed. Our approach, called Monte Carlo Quantization (MCQ), is linear in both time and space, with the resulting quantized, sparse networks showing minimal accuracy loss when compared to the original full-precision networks. Our method either outperforms or achieves competitive results on multiple benchmarks compared to previous quantization methods that do require additional training.

Gonçalo Mordido, Haojin Yang, Christoph Meinel

We propose to incorporate adversarial dropout in generative multi-adversarial networks, by omitting or dropping out, the feedback of each discriminator in the framework with some probability at the end of each batch. Our approach forces the single generator not to constrain its output to satisfy a single discriminator, but, instead, to satisfy a dynamic ensemble of discriminators. We show that this leads to a more generalized generator, promoting variety in the generated samples and avoiding the common mode collapse problem commonly experienced with generative adversarial networks (GANs). We further provide evidence that the proposed framework, named Dropout-GAN, promotes sample diversity both within and across epochs, eliminating mode collapse and stabilizing training.

Gonçalo Mordido, João Magalhães, Sofia Cavaco

Using solely the information retrieved by audio fingerprinting techniques, we propose methods to treat a possibly large dataset of user-generated audio content, that (1) enable the grouping of several audio files that contain a common audio excerpt (i.e., are relative to the same event), and (2) give information about how those files are correlated in terms of time and quality inside each event. Furthermore, we use supervised learning to detect incorrect matches that may arise from the audio fingerprinting algorithm itself, whilst ensuring our model learns with previous predictions. All the presented methods were further validated by user-generated recordings of several different concerts manually crawled from YouTube.

Gonçalo Mordido, João Magalhães, Sofia Cavaco

The increase of the quantity of user-generated content experienced in social media has boosted the importance of analysing and organising the content by its quality. Here, we propose a method that uses audio fingerprinting to organise and infer the quality of user-generated audio content. The proposed method detects the overlapping segments between different audio clips to organise and cluster the data according to events, and to infer the audio quality of the samples. A test setup with concert recordings manually crawled from YouTube is used to validate the presented method. The results show that the proposed method achieves better results than previous methods.