With the popularity of Machine Learning (ML) solutions, algorithms and data have been released faster than the capacity of processing them. In this context, the problem of Algorithm Recommendation (AR) is receiving a significant deal of attention recently. This problem has been addressed in the literature as a learning task, often as a Meta-Learning problem where the aim is to recommend the best alternative for a specific dataset. For such, datasets encoded by meta-features are explored by ML algorithms that try to learn the mapping between meta-representations and the best technique to be used. One of the challenges for the successful use of ML is to define which features are the most valuable for a specific dataset since several meta-features can be used, which increases the meta-feature dimension. This paper presents an empirical analysis of Feature Selection and Feature Extraction in the meta-level for the AR problem. The present study was focused on three criteria: predictive performance, dimensionality reduction, and pipeline runtime. As we verified, applying Dimensionality Reduction (DR) methods did not improve predictive performances in general. However, DR solutions reduced about 80% of the meta-features, obtaining pretty much the same performance as the original setup but with lower runtimes. The only exception was PCA, which presented about the same runtime as the original meta-features. Experimental results also showed that various datasets have many non-informative meta-features and that it is possible to obtain high predictive performance using around 20% of the original meta-features. Therefore, due to their natural trend for high dimensionality, DR methods should be used for Meta-Feature Selection and Meta-Feature Extraction.

The knowledge of a deep learning model may be transferred to a student model, leading to intellectual property infringement or vulnerability propagation. Detecting such knowledge reuse is nontrivial because the suspect models may not be white-box accessible and/or may serve different tasks. In this paper, we propose ModelDiff, a testing-based approach to deep learning model similarity comparison. Instead of directly comparing the weights, activations, or outputs of two models, we compare their behavioral patterns on the same set of test inputs. Specifically, the behavioral pattern of a model is represented as a decision distance vector (DDV), in which each element is the distance between the model's reactions to a pair of inputs. The knowledge similarity between two models is measured with the cosine similarity between their DDVs. To evaluate ModelDiff, we created a benchmark that contains 144 pairs of models that cover most popular model reuse methods, including transfer learning, model compression, and model stealing. Our method achieved 91.7% correctness on the benchmark, which demonstrates the effectiveness of using ModelDiff for model reuse detection. A study on mobile deep learning apps has shown the feasibility of ModelDiff on real-world models.

Bayesian persuasion studies how an informed sender should partially disclose information to influence the behavior of a self-interested receiver. Classical models make the stringent assumption that the sender knows the receiver's utility. This can be relaxed by considering an online learning framework in which the sender repeatedly faces a receiver of an unknown, adversarially selected type. We study, for the first time, an online Bayesian persuasion setting with multiple receivers. We focus on the case with no externalities and binary actions, as customary in offline models. Our goal is to design no-regret algorithms for the sender with polynomial per-iteration running time. First, we prove a negative result: for any $0 < \alpha \leq 1$, there is no polynomial-time no-$\alpha$-regret algorithm when the sender's utility function is supermodular or anonymous. Then, we focus on the case of submodular sender's utility functions and we show that, in this case, it is possible to design a polynomial-time no-$(1 - \frac{1}{e})$-regret algorithm. To do so, we introduce a general online gradient descent scheme to handle online learning problems with a finite number of possible loss functions. This requires the existence of an approximate projection oracle. We show that, in our setting, there exists one such projection oracle which can be implemented in polynomial time.

Graph Neural Networks (GNNs) have been widely applied to various fields due to their powerful representations of graph-structured data. Despite the success of GNNs, most existing GNNs are designed to learn node representations on the fixed and homogeneous graphs. The limitations especially become problematic when learning representations on a misspecified graph or a heterogeneous graph that consists of various types of nodes and edges. To address this limitations, we propose Graph Transformer Networks (GTNs) that are capable of generating new graph structures, which preclude noisy connections and include useful connections (e.g., meta-paths) for tasks, while learning effective node representations on the new graphs in an end-to-end fashion. We further propose enhanced version of GTNs, Fast Graph Transformer Networks (FastGTNs), that improve scalability of graph transformations. Compared to GTNs, FastGTNs are 230x faster and use 100x less memory while allowing the identical graph transformations as GTNs. In addition, we extend graph transformations to the semantic proximity of nodes allowing non-local operations beyond meta-paths. Extensive experiments on both homogeneous graphs and heterogeneous graphs show that GTNs and FastGTNs with non-local operations achieve the state-of-the-art performance for node classification tasks. The code is available: https://github.com/seongjunyun/Graph_Transformer_Networks

Models trained with offline data often suffer from continual distribution shifts and expensive labeling in changing environments. This calls for a new online learning paradigm where the learner can continually adapt to changing environments with limited labels. In this paper, we propose a new online setting -- Online Active Continual Adaptation, where the learner aims to continually adapt to changing distributions using both unlabeled samples and active queries of limited labels. To this end, we propose Online Self-Adaptive Mirror Descent (OSAMD), which adopts an online teacher-student structure to enable online self-training from unlabeled data, and a margin-based criterion that decides whether to query the labels to track changing distributions. Theoretically, we show that, in the separable case, OSAMD has an $O({T}^{1/2})$ dynamic regret bound under mild assumptions, which is even tighter than the lower bound $\Omega(T^{2/3})$ of traditional online learning with full labels. In the general case, we show a regret bound of $O({\alpha^*}^{1/3} {T}^{2/3} + \alpha^* T)$, where $\alpha^*$ denotes the separability of domains and is usually small. Our theoretical results show that OSAMD can fast adapt to changing environments with active queries. Empirically, we demonstrate that OSAMD achieves favorable regrets under changing environments with limited labels on both simulated and real-world data, which corroborates our theoretical findings.

Deep learning empirically achieves high performance in many applications, but its training dynamics has not been fully understood theoretically. In this paper, we explore theoretical analysis on training two-layer ReLU neural networks in a teacher-student regression model, in which a student network learns an unknown teacher network through its outputs. We show that with a specific regularization and sufficient over-parameterization, the student network can identify the parameters of the teacher network with high probability via gradient descent with a norm dependent stepsize even though the objective function is highly non-convex. The key theoretical tool is the measure representation of the neural networks and a novel application of a dual certificate argument for sparse estimation on a measure space. We analyze the global minima and global convergence property in the measure space.

This course introduces machine learning covering the three main techniques used in industry: regression, classification, and clustering. It is designed to be self-contained, easy to approach, and fast to assimilate. The course is designed to maximize the learning experience for everyone and includes 50% theory and 50% hands-on practice. It includes labs with hands-on exercises and solutions. You can run the code on Google CoLab and get started right away.

Welcome to the best Natural Language Processing course on the Udemy! This course is designed to be your complete online resource for learning how to use Natural Language Processing with the Python programming language. In the course we will cover everything you need to learn in order to become a world class practitioner of NLP with Python. We'll start off with the basics, learning how to open and work with text, as well as learning how to use regular expressions to search for custom patterns inside of text files. Afterwards we will begin with the basics of Natural Language Processing, utilizing the Natural Language Toolkit library for Python, as well as the state of the art Spacy library for ultra fast tokenization, parsing, entity recognition, and lemmatization of text.

Training machine learning model for production use is a hectic and time-consuming process. With MLOps, this narrative is changing. MLOps, a descendent of DevOps, provides the automation and scalability required to develop, train, and continuously deliver modern agile machine learning applications.Machine learning lifecycle management tools are important to implement DevOps practices in your machine learning environment. Using popular, efficient open-source tools such as those mentioned in this article is an excellent start to your machine learning MLOps journey.

The EPSRC UK Robotics & Autonomous Systems (UK-RAS Network) has announced the inaugural programme for the UK Festival of Robotics – a brand new 7-day celebration set to boost public engagement in robotics and intelligent systems. This week-long programme will run from 19th – 25th June 2021, offering online events, special competitions, and interactive activities for robot enthusiasts of all ages. Full details of all the events and activities on offer throughout the week are available here: https://www.ukras.org/robotics-festival/ Hosted by Dr Claire Asher and featuring Dr Sophie Armanini (TU Munich/Imperial College London), Matthew Ryan Tucker (University of Bristol) and Ben Scott-Robinson (Small Robot Company), Episode 9 will focus on "Robots into the Wild". In this fun, interactive session, participants will take on the role of a robot accident investigator, work together with others and have fun'solving' the case!