Machine learning involves the use of machine learning algorithms and models. For beginners, this is very confusing as often "machine learning algorithm" is used interchangeably with "machine learning model." Are they the same thing or something different? As a developer, your intuition with "algorithms" like sort algorithms and search algorithms will help to clear up this confusion. In this post, you will discover the difference between machine learning "algorithms" and "models."

A recent source of concern for the security of neural networks is the emergence of clean-label dataset poisoning attacks, wherein correctly labeled poisoned samples are injected in the training dataset. While these poisons look legitimate to the human observer, they contain malicious characteristics that trigger a targeted misclassification during inference. We propose a scalable and transferable clean-label attack, Bullseye Polytope, which creates poison images centered around the target image in the feature space. Bullseye Polytope improves the attack success rate of the current state-of-the-art by 26.75% in end-to-end training, while increasing attack speed by a factor of 12. We further extend Bullseye Polytope to a more practical attack model by including multiple images of the same object (e.g., from different angles) in crafting the poisoned samples. We demonstrate that this extension improves attack transferability by over 16% to unseen images (of the same object) without increasing the number of poisons.

At the heart of machine learning lies the question of generalizability of learned rules over previously unseen data. While over-parameterized models based on neural networks are now ubiquitous in machine learning applications, our understanding of their generalization capabilities is incomplete. This task is made harder by the non-convexity of the underlying learning problems. We provide a general framework to characterize the asymptotic generalization error for single-layer neural networks (i.e., generalized linear models) with arbitrary non-linearities, making it applicable to regression as well as classification problems. This framework enables analyzing the effect of (i) over-parameterization and non-linearity during modeling; and (ii) choices of loss function, initialization, and regularizer during learning. Our model also captures mismatch between training and test distributions. As examples, we analyze a few special cases, namely linear regression and logistic regression. We are also able to rigorously and analytically explain the \emph{double descent} phenomenon in generalized linear models.

While probing is a common technique for identifying knowledge in the representations of pretrained models, it is unclear whether this technique can explain the downstream success of models like BERT which are trained end-to-end during finetuning. To address this question, we compare probing with a different measure of transferability: the decrease in finetuning performance of a partially-reinitialized model. This technique reveals that in BERT, layers with high probing accuracy on downstream GLUE tasks are neither necessary nor sufficient for high accuracy on those tasks. In addition, dataset size impacts layer transferability: the less finetuning data one has, the more important the middle and later layers of BERT become. Furthermore, BERT does not simply find a better initializer for individual layers; instead, interactions between layers matter and reordering BERT's layers prior to finetuning significantly harms evaluation metrics. These results provide a way of understanding the transferability of parameters in pretrained language models, revealing the fluidity and complexity of transfer learning in these models.

Recently machine learning techniques have become popular for analysing physical systems and solving problems occurring in quantum computing. In this paper we focus on using such techniques for finding the sequence of physical operations implementing the given quantum logical operation. In this context we analyse the flexibility of the data representation and compare the applicability of two machine learning approaches based on different representations of data. We demonstrate that the utilization of the geometrical structure of control pulses is sufficient for achieving high-fidelity of the implemented evolution. We also demonstrate that artificial neural networks, unlike geometrical methods, posses the generalization abilities enabling them to generate control pulses for the systems with variable strength of the disturbance. The presented results suggest that in some quantum control scenarios, geometrical data representation and processing is competitive to more complex methods.

The curse of dimensionality causes the well-known and widely discussed problems for machine learning methods. There is a hypothesis that using of the Manhattan distance and even fractional quasinorms lp (for p less than 1) can help to overcome the curse of dimensionality in classification problems. In this study, we systematically test this hypothesis. We confirm that fractional quasinorms have a greater relative contrast or coefficient of variation than the Euclidean norm l2, but we also demonstrate that the distance concentration shows qualitatively the same behaviour for all tested norms and quasinorms and the difference between them decays as dimension tends to infinity. Estimation of classification quality for kNN based on different norms and quasinorms shows that a greater relative contrast does not mean better classifier performance and the worst performance for different databases was shown by different norms (quasinorms). A systematic comparison shows that the difference of the performance of kNN based on lp for p=2, 1, and 0.5 is statistically insignificant.

Equal Opportunity and Fairness are receiving increasing attention in artificial intelligence. Stereotyping is another source of discrimination, which yet has been unstudied in literature. GAN-made faces would be exposed to such discrimination, if they are classified by human perception. It is possible to eliminate the human impact on fictitious faces classification task by the use of statistical approaches. We present a novel approach through penalized regression to label stereotype-free GAN-generated synthetic unlabeled images. The proposed approach aids labeling new data (fictitious output images) by minimizing a penalized version of the least squares cost function between realistic pictures and target pictures.

How many statistical inference tools we have for inference from massive data? A huge number, but only when we are ready to assume the given database is homogenous, consisting of a large cohort of "similar" cases. Why we need the homogeneity assumption? To make `learning from the experience of others' or `borrowing strength' possible. But, what if, we are dealing with a massive database of heterogeneous cases (which is a norm in almost all modern data-science applications including neuroscience, genomics, healthcare, and astronomy)? How many methods we have in this situation? Not much, if not ZERO. Why? It's not obvious how to go about gathering strength when each piece of information is fuzzy. The danger is that, if we include irrelevant cases, borrowing information might heavily damage the quality of the inference! This raises some fundamental questions for big data inference: When (not) to borrow? Whom (not) to borrow? How (not) to borrow? These questions are at the heart of the "Problem of Relevance" in statistical inference -- a puzzle that has remained too little addressed since its inception nearly half a century ago. Here we offer the first practical theory of relevance with precisely describable statistical formulation and algorithm. Through examples, we demonstrate how our new statistical perspective answers previously unanswerable questions in a realistic and feasible way.

Traffic prediction plays an essential role in intelligent transportation system. Accurate traffic prediction can assist route planing, guide vehicle dispatching, and mitigate traffic congestion. This problem is challenging due to the complicated and dynamic spatio-temporal dependencies between different regions in the road network. Recently, a significant amount of research efforts have been devoted to this area, greatly advancing traffic prediction abilities. The purpose of this paper is to provide a comprehensive survey for traffic prediction. Specifically, we first summarize the existing traffic prediction methods, and give a taxonomy of them. Second, we list the common applications of traffic prediction and the state-of-the-art in these applications. Third, we collect and organize widely used public datasets in the existing literature. Furthermore, we give an evaluation by conducting extensive experiments to compare the performance of methods related to traffic demand and speed prediction respectively on two datasets. Finally, we discuss potential future directions.

Time-critical analysis of social media streams is important for humanitarian organizations to plan rapid response during disasters. The crisis informatics research community has developed several techniques and systems to process and classify big crisis related data posted on social media. However, due to the dispersed nature of the datasets used in the literature, it is not possible to compare the results and measure the progress made towards better models for crisis informatics. In this work, we attempt to bridge this gap by standardizing various existing crisis-related datasets. We consolidate labels of eight annotated data sources and provide 166.1k and 141.5k tweets for informativeness and humanitarian classification tasks, respectively. The consolidation results in a larger dataset that affords the ability to train more sophisticated models. To that end, we provide baseline results using CNN and BERT models.