It is the year 2035. Your self-driving car has dropped you in front of your office and your robot receptionist is giving you a list of all visitors. You are hungry and are checking your food (hydroponic plants for fruits and vegetables and in-vitro cloned meat) source on a vertical building plant on a mind-controlled phone. You unload your package, sign the delivery ledger and walk back to your AI controlled cabin that has everything set as per your preferences. No, this is not fiction we are talking about. It is the route that our world has already taken.

Machine learning (ML) and artificial intelligence (AI) algorithms are now being used to automate the discovery of physics principles and governing equations from measurement data alone. However, positing a universal physical law from data is challenging without simultaneously proposing an accompanying discrepancy model to account for the inevitable mismatch between theory and measurements. By revisiting the classic problem of modeling falling objects of different size and mass, we highlight a number of subtle and nuanced issues that must be addressed by modern data-driven methods for the automated discovery of physics. Specifically, we show that measurement noise and complex secondary physical mechanisms, such as unsteady fluid drag forces, can obscure the underlying law of gravitation, leading to an erroneous model. Without proposing an appropriate discrepancy model to handle these drag forces, the data supports an Aristotelian, versus a Galilean, theory of gravitation. Using the sparse identification of nonlinear dynamics (SINDy) algorithm, with the additional assumption that each separate falling object is governed by the same physical law, we are able to identify a viable discrepancy model to account for the fluid dynamic forces that explain the mismatch between a posited universal law of gravity and the measurement data. This work highlights the fact that the simple application of ML/AI will generally be insufficient to extract universal physical laws without further modification.

This paper provides a comprehensive survey of Machine Learning Testing (ML testing) research. It covers 128 papers on testing properties (e.g., correctness, robustness, and fairness), testing components (e.g., the data, learning program, and framework), testing workflow (e.g., test generation and test evaluation), and application scenarios (e.g., autonomous driving, machine translation). The paper also analyses trends concerning datasets, research trends, and research focus, concluding with research challenges and promising research directions in ML testing.

Shannon's seminal theory of information has been of paramount importance in the development of modern machine learning techniques. However, standard information measures deal with probability distributions over an alphabet considered as a mere set of symbols and disregard further geometric structure, which might be available in the form of a metric or similarity function. We advocate the use of a notion of entropy that reflects not only the relative abundances of symbols but also the similarities between them, which was originally introduced in theoretical ecology to study the diversity of biological communities. Echoing this idea, we propose a criterion for comparing two probability distributions (possibly degenerate and with non-overlapping supports) that takes into account the geometry of the space in which the distributions are defined. Our proposal exhibits performance on par with state-of-the-art methods based on entropy-regularized optimal transport, but enjoys a closed-form expression and thus a lower computational cost. We demonstrate the versatility of our proposal via experiments on a broad range of domains: computing image barycenters, approximating densities with a collection of (super-) samples; summarizing texts; assessing mode coverage; as well as training generative models.

Recent advances in Neural Architecture Search (NAS) have produced state-of-the-art architectures on several tasks. NAS shifts the efforts of human experts from developing novel architectures directly to designing architecture search spaces and methods to explore them efficiently. The search space definition captures prior knowledge about the properties of the architectures and it is crucial for the complexity and the performance of the search algorithm. However, different search space definitions require restarting the learning process from scratch. We propose a novel agent based on the Transformer that supports joint training and efficient transfer of prior knowledge between multiple search spaces and tasks.

With the prevalence of machine learning services, crowdsourced data containing sensitive information poses substantial privacy challenges. Existing works focusing on protecting against membership inference attacks under the rigorous notion of differential privacy are susceptible to attribute inference attacks. In this paper, we develop a theoretical framework for task-specific privacy under the attack of attribute inference. Under our framework, we propose a minimax optimization formulation with a practical algorithm to protect a given attribute and preserve utility. We also extend our formulation so that multiple attributes could be simultaneously protected. Theoretically, we prove an information-theoretic lower bound to characterize the inherent tradeoff between utility and privacy when they are correlated. Empirically, we conduct experiments with real-world tasks that demonstrate the effectiveness of our method compared with state-of-the-art baseline approaches.

This paper introduces a novel optimization method for differential neural architecture search, based on the theory of prediction with expert advice. Its optimization criterion is well fitted for an architecture-selection, i.e., it minimizes the regret incurred by a sub-optimal selection of operations. Unlike previous search relaxations, that require hard pruning of architectures, our method is designed to dynamically wipe out inferior architectures and enhance superior ones. It achieves an optimal worst-case regret bound and suggests the use of multiple learning-rates, based on the amount of information carried by the backward gradients. Experiments show that our algorithm achieves a strong performance over several image classification datasets. Specifically, it obtains an error rate of 1.6% for CIFAR-10, 24% for ImageNet under mobile settings, and achieves state-of-the-art results on three additional datasets.

Multi-view clustering is an important and fundamental problem. Many multi-view subspace clustering methods have been proposed and achieved success in real-world applications, most of which assume that all views share a same coefficient matrix. However, the underlying information of multiview data are not exploited effectively under this assumption, since the coefficient matrices of different views should have the same clustering properties rather than be the same among multiple views. To this end, a novel Constrained Bilinear Factorization Multi-view Subspace Clustering (CBF-MSC) method is proposed in this paper. Specifically, the bilinear factorization with an orthonormality constraint and a low-rank constraint is employed for all coefficient matrices to make all coefficient matrices have the same trace-norm instead of being equivalent, so as to explore the consensus information of multi-view data more effectively. Finally, an algorithm based on the Augmented Lagrangian Multiplier (ALM) scheme with alternating direction minimization is designed to optimize the objective function. Comprehensive experiments tested on six benchmark datasets validate the effectiveness and competitiveness of the proposed approach compared with several state-of-the-art approaches.

We consider supervised dimension reduction problems, namely to identify a low dimensional projection of the predictors $\-x$ which can retain the statistical relationship between $\-x$ and the response variable $y$. We follow the idea of the sliced inverse regression (SIR) class of methods, which is to use the statistical information of the conditional distribution $\pi(\-x|y)$ to identify the dimension reduction (DR) space and in particular we focus on the task of computing this conditional distribution. We propose a Bayesian framework to compute the conditional distribution where the likelihood function is obtained using the Gaussian process regression model. The conditional distribution $\pi(\-x|y)$ can then be obtained directly by assigning weights to the original data points. We then can perform DR by considering certain moment functions (e.g. the first moment) of the samples of the posterior distribution. With numerical examples, we demonstrate that the proposed method is especially effective for small data problems.

We present a novel intrinsically motivated agent that learns how to control the environment in the fastest possible manner by optimizing learning progress. It learns what can be controlled, how to allocate time and attention, and the relations between objects using surprise based motivation. The effectiveness of our method is demonstrated in a synthetic as well as a robotic manipulation environment yielding considerably improved performance and smaller sample complexity. In a nutshell, our work combines several task-level planning agent structures (backtracking search on task graph, probabilistic road-maps, allocation of search efforts) with intrinsic motivation to achieve learning from scratch.