Computer simulations have become a popular tool of assessing complex skills such as problem-solving skills. Log files of computer-based items record the entire human-computer interactive processes for each respondent. The response processes are very diverse, noisy, and of nonstandard formats. Few generic methods have been developed for exploiting the information contained in process data. In this article, we propose a method to extract latent variables from process data. The method utilizes a sequence-to-sequence autoencoder to compress response processes into standard numerical vectors. It does not require prior knowledge of the specific items and human-computers interaction patterns. The proposed method is applied to both simulated and real process data to demonstrate that the resulting latent variables extract useful information from the response processes.

We propose a method to open the black box of the Multi-Layer Perceptron by inferring from it a simpler and generally more accurate general additive model. The resulting model comprises non-linear univariate and bivariate partial responses derived from the original Multi-Layer Perceptron. The responses are combined using the Lasso and further optimised within a modular structure. The approach is generic and provides a constructive framework to simplify and explain the Multi-Layer Perceptron for any data set, opening the door for validation against prior knowledge. Experimental results on benchmarking datasets indicate that the partial responses are intuitive to interpret and the Area Under the Curve is competitive with Gradient Boosting, Support Vector Machines and Random Forests. The performance improvement compared with a fully connected Multi-Layer Perceptron is attributed to reduced confounding in the second stage of optimisation of the weights. The main limitation of the method is that it explicitly models only up to pairwise interactions. For many practical applications this will be optimal, but where that is not the case then this will be indicated by the performance difference compared to the original model. The streamlined model simultaneously interprets and optimises this frequently used flexible model.

These pro-3 grams rely on patterns and inference, rather than explicit instructions, to achieve their aims [5]. ML in construction has been developed significantly since 1991 when [6] first discussed the potential of neural networks in construction engineering and management. Early examples of ML in construction include applications such as [7] where the AQ15 algorithm was applied to automatically learn the mapping between constructability (poor, good, excellent) and 7 predictors from a collection of 31 training examples; and [8] who applied decision trees and neural networks to a construction management database to identify the causes of delays. Many subsequent prediction applications applied support vector machines (SVMs), owing to their consistently high accuracy. These applications include [9], who accurately forecasted contractor prequalification using input variables such as financial strength and current workload; [10], who estimated building cost and loss risk from ten input variables; and [11], who detected concrete structural components in color images from actual construction sites. In the last 5 years, use of ML in construction has become far more widespread and the methods and applications used are far more diverse.

Supervised learning algorithms trained on medical images will often fail to generalize across changes in acquisition parameters. Recent work in domain adaptation addresses this challenge and successfully leverages labeled data in a source domain to perform well on an unlabeled target domain. Inspired by recent work in semi-supervised learning we introduce a novel method to adapt from one source domain to $n$ target domains (as long as there is paired data covering all domains). Our multi-domain adaptation method utilises a consistency loss combined with adversarial learning. We provide results on white matter lesion hyperintensity segmentation from brain MRIs using the MICCAI 2017 challenge data as the source domain and two target domains. The proposed method significantly outperforms other domain adaptation baselines.

Federated learning (FL) enables on-device training over distributed networks consisting of a massive amount of modern smart devices, such as smartphones and IoT devices. However, the leading optimization algorithm in such settings, i.e., federated averaging (FedAvg), suffers from heavy communication cost and inevitable performance drop, especially when the local data is distributed in a non-IID way. To alleviate this problem, we propose two potential solutions by introducing additional mechanisms to the on-device training. The first (FedMMD) is adopting a two-stream model with the MMD (Maximum Mean Discrepancy) constraint instead of a single model in vanilla FedAvg to be trained on devices. Experiments show that the proposed method outperforms baselines, especially in non-IID FL settings, with a reduction of more than 20% in required communication rounds. The second is FL with feature fusion (FedFusion). By aggregating the features from both the local and global models, we achieve higher accuracy at less communication cost. Furthermore, the feature fusion modules offer better initialization for newly incoming clients and thus speed up the process of convergence. Experiments in popular FL scenarios show that our FedFusion outperforms baselines in both accuracy and generalization ability while reducing the number of required communication rounds by more than 60%.

In biomarker studies it is popular to perform an unsupervised clustering of high-dimensional variables like genome wide screens of SNPs, gene expressions, and protein data and regress for example treatment response, patient recorded outcome measures, time to disease progression, or overall survival on these potentially mislabelled clusters. It is well-known from the statistical literature that errors in continuous and categorical covariates can lead to loss of important information about effects on outcome (Carroll et al., 2006). However, to our surprise this is often ignored when regressing outcome on classes identified by unsupervised learning, which might lead to important clinical effect measures being overlooked (Alizadeh et al., 2000; Veer et al., 2002; Guinney et al., 2015; Zhan et al., 2006; Broyl et al., 2010). We suggest to cast the problem as a covariate misclassification problem. This leaves us with a concourse of possible modelling and analysis options, see for example the book by Carroll et al. (2006) or the recent review by Brakenhoff et al. (2018).

It is well-known that overparametrized neural networks trained using gradient-based methods quickly achieve small training error with appropriate hyperparameter settings. Recent papers have proved this statement theoretically for highly overparametrized networks under reasonable assumptions. The limiting case when the network size approaches infinity has also been considered. These results either assume that the activation function is ReLU or they crucially depend on the minimum eigenvalue of a certain Gram matrix depending on the data, random initialization and the activation function. In the latter case, existing works only prove that this minimum eigenvalue is non-zero and do not provide quantitative bounds. On the empirical side, a contemporary line of investigations has proposed a number of alternative activation functions which tend to perform better than ReLU at least in some settings but no clear understanding has emerged. This state of affairs underscores the importance of theoretically understanding the impact of activation functions on training. In the present paper, we provide theoretical results about the effect of activation function on the training of highly overparametrized 2-layer neural networks. We show that for smooth activations, such as tanh and swish, the minimum eigenvalue can be exponentially small depending on the span of the dataset implying that the training can be very slow. In contrast, for activations with a "kink," such as ReLU, SELU, ELU, all eigenvalues are large under minimal assumptions on the data. Several new ideas are involved. Finally, we corroborate our results empirically.

Active learning (AL) is a widely-used training strategy for maximizing predictive performance subject to a fixed annotation budget. In AL one iteratively selects training examples for annotation, often those for which the current model is most uncertain (by some measure). The hope is that active sampling leads to better performance than would be achieved under independent and identically distributed (i.i.d.) random samples. While AL has shown promise in retrospective evaluations, these studies often ignore practical obstacles to its use. In this paper we show that while AL may provide benefits when used with specific models and for particular domains, the benefits of current approaches do not generalize reliably across models and tasks. This is problematic because in practice one does not have the opportunity to explore and compare alternative AL strategies. Moreover, AL couples the training dataset with the model used to guide its acquisition. We find that subsequently training a successor model with an actively-acquired dataset does not consistently outperform training on i.i.d. sampled data. Our findings raise the question of whether the downsides inherent to AL are worth the modest and inconsistent performance gains it tends to afford.

Continuous Relaxation of Symbolic Planner for One-Shot Imitation Learning De-An Huang 1, Danfei Xu 1, Y uke Zhu 1, Animesh Garg 1, 2, Silvio Savarese 1, Li Fei-Fei 1, Juan Carlos Niebles 1 Abstract -- We address one-shot imitation learning, where the goal is to execute a previously unseen task based on a single demonstration. While there has been exciting progress in this direction, most of the approaches still require a few hundred tasks for meta-training, which limits the scalability of the approaches. Our main contribution is to formulate one-shot imitation learning as a symbolic planning problem along with the symbol grounding problem. This formulation disentangles the policy execution from the inter-task generalization and leads to better data efficiency. The key technical challenge is that the symbol grounding is prone to error with limited training data and leads to subsequent symbolic planning failures. We address this challenge by proposing a continuous relaxation of the discrete symbolic planner that directly plans on the probabilistic outputs of the symbol grounding model. Our continuous relaxation of the planner can still leverage the information contained in the probabilistic symbol grounding and significantly improve over the baseline planner for the one-shot imitation learning tasks without using large training data. I NTRODUCTION We are interested in robots that can learn a wide variety of tasks efficiently. Recently, there has been an increasing interest in the one-shot imitation learning problem [1-7], where the goal is to learn to execute a previously unseen task from only a single demonstration of the task. This setting is also referred to as meta-learning [3, 8], where the meta-training stage uses a set of tasks in a given domain to simulate the one-shot testing scenario. This allows the learned model to generalize to previously unseen tasks with a single demonstration in the meta-testing stage. The main shortcoming of these one-shot approaches is that they typically require a large amount of data for meta-training (400 meta-training tasks in [4] and 1000 in [6] for the Block Stacking task [6]) to generalize reliably to unseen tasks.

Multi-agent systems have a wide range of applications in cooperative and competitive tasks. As the number of agents increases, nonstationarity gets more serious in multi-agent reinforcement learning (MARL), which brings great difficulties to the learning process. Besides, current mainstream algorithms configure each agent an independent network,so that the memory usage increases linearly with the number of agents which greatly slows down the interaction with the environment. Inspired by Generative Adversarial Networks (GAN), this paper proposes an iterative update method (IU) to stabilize the nonstationary environment. Further, we add first-person perspective and represent all agents by only one network which can change agents' policies from sequential compute to batch compute. Similar to continual lifelong learning, we realize the iterative update method in this unified representative network (IUUR). In this method, iterative update can greatly alleviate the nonstationarity of the environment, unified representation can speed up the interaction with environment and avoid the linear growth of memory usage. Besides, this method does not bother decentralized execution and distributed deployment. Experiments show that compared with MADDPG, our algorithm achieves state-of-the-art performance and saves wall-clock time by a large margin especially with more agents.