Word embeddings, i.e., semantically meaningful vector representation of words, are largely influenced by the distributional hypothesis "You shall know a word by the company it keeps" (Harris, 1954), whereas modern prediction-based neural network embeddings rely on design choices and hyperparameter optimization. Word embeddings like Word2Vec, GloVe etc. well capture the contextuality and real-world analogies but contemporary convolution-based image embeddings such as VGGNet, AlexNet, etc. do not capture contextual knowledge. The popular king-queen analogy does not hold true for most commonly used vision embeddings. In this paper, we introduce a pre-trained joint embedding (JE), named IMAGINATOR, trained on 21K distinct image objects level from 1M image+text pairs. JE is a way to encode multimodal data into a vector space where the text modality serves as the ground-ing key, which the complementary modality (in this case, the image) is anchored with. IMAGINATOR encapsulates three individual representations: (i) object-object co-location, (ii) word-object co-location, and (iii) word-object correlation. These three ways capture complementary aspects of the two modalities which are further combined to obtain the final JEs. Generated JEs are intrinsically evaluated to assess how well they capture the contextuality and real-world analogies. We also evaluate pre-trained IMAGINATOR JEs on three downstream tasks: (i) image captioning, (ii) Image2Tweet, and (iii) text-based image retrieval. IMAGINATOR establishes a new standard on the aforementioned down-stream tasks by outperforming the current SoTA on all the selected tasks. IMAGINATOR will be made publicly available. The codes are available at https://github.com/varunakk/IMAGINATOR
Sparse Neural Networks (SNNs) can potentially demonstrate similar performance to their dense counterparts while saving significant energy and memory at inference. However, the accuracy drop incurred by SNNs, especially at high pruning ratios, can be an issue in critical deployment conditions. While recent works mitigate this issue through sophisticated pruning techniques, we shift our focus to an overlooked factor: hyperparameters and activation functions. Our analyses have shown that the accuracy drop can additionally be attributed to (i) Using ReLU as the default choice for activation functions unanimously, and (ii) Fine-tuning SNNs with the same hyperparameters as dense counterparts. Thus, we focus on learning a novel way to tune activation functions for sparse networks and combining these with a separate hyperparameter optimization (HPO) regime for sparse networks. By conducting experiments on popular DNN models (LeNet-5, VGG-16, ResNet-18, and EfficientNet-B0) trained on MNIST, CIFAR-10, and ImageNet-16 datasets, we show that the novel combination of these two approaches, dubbed Sparse Activation Function Search, short: SAFS, results in up to 15.53%, 8.88%, and 6.33% absolute improvement in the accuracy for LeNet-5, VGG-16, and ResNet-18 over the default training protocols, especially at high pruning ratios. Our code can be found at https://github.com/automl/SAFS
Hyperparameter optimization is crucial to achieving high performance in deep learning. On top of the performance, other criteria such as inference time or memory requirement often need to be optimized due to some practical reasons. This motivates research on multi-objective optimization (MOO). However, Pareto fronts of MOO methods are often shown without considering the variability caused by random seeds and this makes the performance stability evaluation difficult. Although there is a concept named empirical attainment surface to enable the visualization with uncertainty over multiple runs, there is no major Python package for empirical attainment surface. We, therefore, develop a Python package for this purpose and describe the usage. The package is available at https://github.com/nabenabe0928/empirical-attainment-func.
Hyperparameter optimization (HPO) is a powerful technique for automating the tuning of machine learning (ML) models. However, in many real-world applications, accuracy is only one of multiple performance criteria that must be considered. Optimizing these objectives simultaneously on a complex and diverse search space remains a challenging task. In this paper, we propose MO-DEHB, an effective and flexible multi-objective (MO) optimizer that extends the recent evolutionary Hyperband method DEHB. We validate the performance of MO-DEHB using a comprehensive suite of 15 benchmarks consisting of diverse and challenging MO problems, including HPO, neural architecture search (NAS), and joint NAS and HPO, with objectives including accuracy, latency and algorithmic fairness. A comparative study against state-of-the-art MO optimizers demonstrates that MO-DEHB clearly achieves the best performance across our 15 benchmarks.
Many state-of-the-art hyperparameter optimization (HPO) algorithms rely on model-based optimizers that learn surrogate models of the target function to guide the search. Gaussian processes are the de facto surrogate model due to their ability to capture uncertainty but they make strong assumptions about the observation noise, which might not be warranted in practice. In this work, we propose to leverage conformalized quantile regression which makes minimal assumptions about the observation noise and, as a result, models the target function in a more realistic and robust fashion which translates to quicker HPO convergence on empirical benchmarks. To apply our method in a multi-fidelity setting, we propose a simple, yet effective, technique that aggregates observed results across different resource levels and outperforms conventional methods across many empirical tasks.
Well-tuned hyperparameters are crucial for obtaining good generalization behavior in neural networks. They can enforce appropriate inductive biases, regularize the model and improve performance -- especially in the presence of limited data. In this work, we propose a simple and efficient way for optimizing hyperparameters inspired by the marginal likelihood, an optimization objective that requires no validation data. Our method partitions the training data and a neural network model into $K$ data shards and parameter partitions, respectively. Each partition is associated with and optimized only on specific data shards. Combining these partitions into subnetworks allows us to define the ``out-of-training-sample" loss of a subnetwork, i.e., the loss on data shards unseen by the subnetwork, as the objective for hyperparameter optimization. We demonstrate that we can apply this objective to optimize a variety of different hyperparameters in a single training run while being significantly computationally cheaper than alternative methods aiming to optimize the marginal likelihood for neural networks. Lastly, we also focus on optimizing hyperparameters in federated learning, where retraining and cross-validation are particularly challenging.
Gaussian process regression is a well-established Bayesian machine learning method. We propose a new approach to Gaussian process regression using quantum kernels based on parameterized quantum circuits. By employing a hardware-efficient feature map and careful regularization of the Gram matrix, we demonstrate that the variance information of the resulting quantum Gaussian process can be preserved. We also show that quantum Gaussian processes can be used as a surrogate model for Bayesian optimization, a task that critically relies on the variance of the surrogate model. To demonstrate the performance of this quantum Bayesian optimization algorithm, we apply it to the hyperparameter optimization of a machine learning model which performs regression on a real-world dataset. We benchmark the quantum Bayesian optimization against its classical counterpart and show that quantum version can match its performance.
We propose an evolution strategies-based algorithm for estimating gradients in unrolled computation graphs, called ES-Single. Similarly to the recently-proposed Persistent Evolution Strategies (PES), ES-Single is unbiased, and overcomes chaos arising from recursive function applications by smoothing the meta-loss landscape. ES-Single samples a single perturbation per particle, that is kept fixed over the course of an inner problem (e.g., perturbations are not re-sampled for each partial unroll). Compared to PES, ES-Single is simpler to implement and has lower variance: the variance of ES-Single is constant with respect to the number of truncated unrolls, removing a key barrier in applying ES to long inner problems using short truncations. We show that ES-Single is unbiased for quadratic inner problems, and demonstrate empirically that its variance can be substantially lower than that of PES. ES-Single consistently outperforms PES on a variety of tasks, including a synthetic benchmark task, hyperparameter optimization, training recurrent neural networks, and training learned optimizers.
Recent advances in many domains require more and more complicated experiment design. Such complicated experiments often have many parameters, which necessitate parameter tuning. Tree-structured Parzen estimator (TPE), a Bayesian optimization method, is widely used in recent parameter tuning frameworks. Despite its popularity, the roles of each control parameter and the algorithm intuition have not been discussed so far. In this tutorial, we will identify the roles of each control parameter and their impacts on hyperparameter optimization using a diverse set of benchmarks. We compare our recommended setting drawn from the ablation study with baseline methods and demonstrate that our recommended setting improves the performance of TPE. Our TPE implementation is available at https://github.com/nabenabe0928/tpe/tree/single-opt.
This paper proposes a natural evolution strategy (NES) for mixed-integer black-box optimization (MI-BBO) that appears in real-world problems such as hyperparameter optimization of machine learning and materials design. This problem is difficult to optimize because plateaus where the values do not change appear when the integer variables are relaxed to the continuous ones. CMA-ES w. Margin that addresses the plateaus reportedly showed good performance on MI-BBO benchmark problems. However, it has been observed that the search performance of CMA-ES w. Margin deteriorates when continuous variables contribute more to the objective function value than integer ones. In order to address the problem of CMA-ES w. Margin, we propose Distance-weighted eXponential Natural Evolution Strategy taking account of Implicit Constraint and Integer (DX-NES-ICI). We compare the search performance of DX-NES-ICI with that of CMA-ES w. Margin through numerical experiments. As a result, DX-NES-ICI was up to 3.7 times better than CMA-ES w. Margin in terms of a rate of finding the optimal solutions on benchmark problems where continuous variables contribute more to the objective function value than integer ones. DX-NES-ICI also outperformed CMA-ES w. Margin on problems where CMA-ES w. Margin originally showed good performance.
The recent rise in popularity of Hyperparameter Optimization (HPO) for deep learning has highlighted the role that good hyperparameter (HP) space design can play in training strong models. In turn, designing a good HP space is critically dependent on understanding the role of different HPs. This motivates research on HP Importance (HPI), e.g., with the popular method of functional ANOVA (f-ANOVA). However, the original f-ANOVA formulation is inapplicable to the subspaces most relevant to algorithm designers, such as those defined by top performance. To overcome this issue, we derive a novel formulation of f-ANOVA for arbitrary subspaces and propose an algorithm that uses Pearson divergence (PED) to enable a closed-form calculation of HPI. We demonstrate that this new algorithm, dubbed PED-ANOVA, is able to successfully identify important HPs in different subspaces while also being extremely computationally efficient.
In this paper, we introduce a Gaussian process based moving horizon estimation (MHE) framework. The scheme is based on offline collected data and offline hyperparameter optimization. In particular, compared to standard MHE schemes, we replace the mathematical model of the system by the posterior mean of the Gaussian process. To account for the uncertainty of the learned model, we exploit the posterior variance of the learned Gaussian process in the weighting matrices of the cost function of the proposed MHE scheme. We prove practical robust exponential stability of the resulting estimator using a recently proposed Lyapunov-based proof technique. Finally, the performance of the Gaussian process based MHE scheme is illustrated via a nonlinear system.
Numerous research studies in the field of federated learning (FL) have attempted to use personalization to address the heterogeneity among clients, one of FL's most crucial and challenging problems. However, existing works predominantly focus on tailoring models. Yet, due to the heterogeneity of clients, they may each require different choices of hyperparameters, which have not been studied so far. We pinpoint two challenges of personalized federated hyperparameter optimization (pFedHPO): handling the exponentially increased search space and characterizing each client without compromising its data privacy. To overcome them, we propose learning a \textsc{H}yper\textsc{P}arameter \textsc{N}etwork (HPN) fed with client encoding to decide personalized hyperparameters. The client encoding is calculated with a random projection-based procedure to protect each client's privacy. Besides, we design a novel mechanism to debias the low-fidelity function evaluation samples for learning HPN. We conduct extensive experiments on FL tasks from various domains, demonstrating the superiority of HPN.
Although Reinforcement Learning (RL) has shown to be capable of producing impressive results, its use is limited by the impact of its hyperparameters on performance. This often makes it difficult to achieve good results in practice. Automated RL (AutoRL) addresses this difficulty, yet little is known about the dynamics of the hyperparameter landscapes that hyperparameter optimization (HPO) methods traverse in search of optimal configurations. In view of existing AutoRL approaches dynamically adjusting hyperparameter configurations, we propose an approach to build and analyze these hyperparameter landscapes not just for one point in time but at multiple points in time throughout training. Addressing an important open question on the legitimacy of such dynamic AutoRL approaches, we provide thorough empirical evidence that the hyperparameter landscapes strongly vary over time across representative algorithms from RL literature (DQN, PPO, and SAC) in different kinds of environments (Cartpole, Bipedal Walker, and Hopper) This supports the theory that hyperparameters should be dynamically adjusted during training and shows the potential for more insights on AutoRL problems that can be gained through landscape analyses. Our code can be found at https://github.com/automl/AutoRL-Landscape
Neural networks have revolutionized many aspects of society but in the era of huge models with billions of parameters, optimizing and deploying them for commercial applications can require significant computational and financial resources. To address these challenges, we introduce the Tetra-AML toolbox, which automates neural architecture search and hyperparameter optimization via a custom-developed black-box Tensor train Optimization algorithm, TetraOpt. The toolbox also provides model compression through quantization and pruning, augmented by compression using tensor networks. Here, we analyze a unified benchmark for optimizing neural networks in computer vision tasks and show the superior performance of our approach compared to Bayesian optimization on the CIFAR-10 dataset. We also demonstrate the compression of ResNet-18 neural networks, where we use 14.5 times less memory while losing just 3.2% of accuracy. The presented framework is generic, not limited by computer vision problems, supports hardware acceleration (such as with GPUs and TPUs) and can be further extended to quantum hardware and to hybrid quantum machine learning models.
This paper presents a systematic overview of parameter-efficient fine-tuning methods, covering over 50 papers published between early 2019 and mid-2024. These methods aim to address the challenges of fine-tuning large language models by training only a small subset of parameters. We provide a taxonomy that covers a broad range of methods and present a detailed method comparison with a specific focus on real-life efficiency in fine-tuning multibillion-scale language models. We also conduct an extensive head-to-head experimental comparison of 15 diverse PEFT methods, evaluating their performance and efficiency on models up to 11B parameters. Our findings reveal that methods previously shown to surpass a strong LoRA baseline face difficulties in resource-constrained settings, where hyperparameter optimization is limited and the network is fine-tuned only for a few epochs. Finally, we provide a set of practical recommendations for using PEFT methods and outline potential future research directions.
Automatically optimizing the hyperparameters of Machine Learning algorithms is one of the primary open questions in AI. Existing work in Hyperparameter Optimization (HPO) trains surrogate models for approximating the response surface of hyperparameters as a regression task. In contrast, we hypothesize that the optimal strategy for training surrogates is to preserve the ranks of the performances of hyperparameter configurations as a Learning to Rank problem. As a result, we present a novel method that meta-learns neural network surrogates optimized for ranking the configurations' performances while modeling their uncertainty via ensembling. In a large-scale experimental protocol comprising 12 baselines, 16 HPO search spaces and 86 datasets/tasks, we demonstrate that our method achieves new state-of-the-art results in HPO.
Training and Hyperparameter Optimization (HPO) of deep learning-based AI models are often compute resource intensive and calls for the use of large-scale distributed resources as well as scalable and resource efficient hyperparameter search algorithms. This work studies the potential of using model performance prediction to aid the HPO process carried out on High Performance Computing systems. In addition, a quantum annealer is used to train the performance predictor and a method is proposed to overcome some of the problems derived from the current limitations in quantum systems as well as to increase the stability of solutions. This allows for achieving results on a quantum machine comparable to those obtained on a classical machine, showing how quantum computers could be integrated within classical machine learning tuning pipelines. Furthermore, results are presented from the development of a containerized benchmark based on an AI-model for collision event reconstruction that allows us to compare and assess the suitability of different hardware accelerators for training deep neural networks.
Spiking neural networks (SNNs) have gained attention as a promising alternative to traditional artificial neural networks (ANNs) due to their potential for energy efficiency and their ability to model spiking behavior in biological systems. However, the training of SNNs is still a challenging problem, and new techniques are needed to improve their performance. In this paper, we study the impact of skip connections on SNNs and propose a hyperparameter optimization technique that adapts models from ANN to SNN. We demonstrate that optimizing the position, type, and number of skip connections can significantly improve the accuracy and efficiency of SNNs by enabling faster convergence and increasing information flow through the network. Our results show an average +8% accuracy increase on CIFAR-10-DVS and DVS128 Gesture datasets adaptation of multiple state-of-the-art models.
Deep learning-based image registration approaches have shown competitive performance and run-time advantages compared to conventional image registration methods. However, existing learning-based approaches mostly require to train separate models with respect to different regularization hyperparameters for manual hyperparameter searching and often do not allow spatially-variant regularization. In this work, we propose a learning-based registration approach based on a novel conditional spatially adaptive instance normalization (CSAIN) to address these challenges. The proposed method introduces a spatially-variant regularization and learns its effect of achieving spatially-adaptive regularization by conditioning the registration network on the hyperparameter matrix via CSAIN. This allows varying of spatially adaptive regularization at inference to obtain multiple plausible deformations with a single pre-trained model. Additionally, the proposed method enables automatic hyperparameter optimization to avoid manual hyperparameter searching. Experiments show that our proposed method outperforms the baseline approaches while achieving spatially-variant and adaptive regularization.