Chengrun Yang, Jicong Fan, Ziyang Wu, and Madeleine Udell This is an extended version of AutoML Pipeline Selection: Efficiently Navigating the Combinatorial Space (DOI: 10.1145/3394486.3403197) at the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2020. Abstract--Data scientists seeking a good supervised learning model on a new dataset have many choices to make: they must preprocess the data, select features, possibly reduce the dimension, select an estimation algorithm, and choose hyperparameters for each of these pipeline components. With new pipeline components comes a combinatorial explosion in the number of choices! In this work, we design a new AutoML system to address this challenge: an automated system to design a supervised learning pipeline. Our system uses matrix and tensor factorization as surrogate models to model the combinatorial pipeline search space.

Agile control of mobile manipulator is challenging because of the high complexity coupled by the robotic system and the unstructured working environment. Tracking and grasping a dynamic object with a random trajectory is even harder. In this paper, a multi-task reinforcement learning-based mobile manipulation control framework is proposed to achieve general dynamic object tracking and grasping. Several basic types of dynamic trajectories are chosen as the task training set. To improve the policy generalization in practice, random noise and dynamics randomization are introduced during the training process. Extensive experiments show that our policy trained can adapt to unseen random dynamic trajectories with about 0.1m tracking error and 75\% grasping success rate of dynamic objects. The trained policy can also be successfully deployed on a real mobile manipulator.

In order to improve the performance of Bayesian optimisation, we develop a modified Gaussian process upper confidence bound (GP-UCB) acquisition function. This is done by sampling the exploration-exploitation trade-off parameter from a distribution. We prove that this allows the expected trade-off parameter to be altered to better suit the problem without compromising a bound on the function's Bayesian regret. We also provide results showing that our method achieves better performance than GP-UCB in a range of real-world and synthetic problems.

Graph Neural Networks (GNNs) are emerging machine learning models on graphs. Although sufficiently deep GNNs are shown theoretically capable of fully preserving graph structures, most existing GNN models in practice are shallow and essentially feature-centric. We show empirically and analytically that the existing shallow GNNs cannot preserve graph structures well. To overcome this fundamental challenge, we propose Eigen-GNN, a simple yet effective and general plug-in module to boost GNNs ability in preserving graph structures. Specifically, we integrate the eigenspace of graph structures with GNNs by treating GNNs as a type of dimensionality reduction and expanding the initial dimensionality reduction bases. Without needing to increase depths, Eigen-GNN possesses more flexibilities in handling both feature-driven and structure-driven tasks since the initial bases contain both node features and graph structures. We present extensive experimental results to demonstrate the effectiveness of Eigen-GNN for tasks including node classification, link prediction, and graph isomorphism tests.

We invited three industry expert speakers using AI to battle climate change. During the hour long webinar, Anita Faul, Data Scientist at the British Antarctic Survey, Lauren Kuntz, CEO and Co-Founder of Gaiascope and Topher White, CEO and Founder of Rainforest Connections walked us through their business use applications of AI to fight the change in climate. Anita started her talk with an explanation of the Thwaites Glacier, otherwise know as the'Doomsday Glacier'. This glacier is responsible for 4% of all sea level increase - if it were to melt completely, sea levels would rise by half a meter in total (hence the name). Therefore, Anita's objective at the Antarctic Survey is to identify icebergs efficiently and reliably in Synthetics Aperture Radar (SAR) satellite images to estimate ice loss.

We consider the contextual bandit problem, where a player sequentially makes decisions based on past observations to maximize the cumulative reward. Although many algorithms have been proposed for contextual bandit, most of them rely on finding the maximum likelihood estimator at each iteration, which requires $O(t)$ time at the $t$-th iteration and are memory inefficient. A natural way to resolve this problem is to apply online stochastic gradient descent (SGD) so that the per-step time and memory complexity can be reduced to constant with respect to $t$, but a contextual bandit policy based on online SGD updates that balances exploration and exploitation has remained elusive. In this work, we show that online SGD can be applied to the generalized linear bandit problem. The proposed SGD-TS algorithm, which uses a single-step SGD update to exploit past information and uses Thompson Sampling for exploration, achieves $\tilde{O}(\sqrt{dT})$ regret with the total time complexity that scales linearly in $T$ and $d$, where $T$ is the total number of rounds and $d$ is the number of features. Experimental results show that SGD-TS consistently outperforms existing algorithms on both synthetic and real datasets.

In this paper, we introduce proximal gradient temporal difference learning, which provides a principled way of designing and analyzing true stochastic gradient temporal difference learning algorithms. We show how gradient TD (GTD) reinforcement learning methods can be formally derived, not by starting from their original objective functions, as previously attempted, but rather from a primal-dual saddle-point objective function. We also conduct a saddle-point error analysis to obtain finite-sample bounds on their performance. Previous analyses of this class of algorithms use stochastic approximation techniques to prove asymptotic convergence, and do not provide any finite-sample analysis. We also propose an accelerated algorithm, called GTD2-MP, that uses proximal ``mirror maps'' to yield an improved convergence rate. The results of our theoretical analysis imply that the GTD family of algorithms are comparable and may indeed be preferred over existing least squares TD methods for off-policy learning, due to their linear complexity. We provide experimental results showing the improved performance of our accelerated gradient TD methods.

Ever since the 1930s when scientists, namely Hans Bethe, discovered that nuclear fusion was possible, researchers strived to initiate and control fusion reactions to produce useful energy on Earth. The best example of a fusion reaction is in the middle of stars like the Sun where hydrogen atoms are fused together to make helium releasing a lot of energy that powers the heat and light of the star. On Earth, scientists need to heat and control plasma, an ionised state of matter similar to gas, to cause particles to fuse and release their energy. Unfortunately, it is very difficult to start fusion reactions on Earth, as they require conditions similar to the Sun, very high temperature and pressure, and scientists have been trying to find a solution for decades. In May 2019, a workshop detailing how fusion could be advanced using machine learning was held that was jointly supported by the Department of Energy Offices of Fusion Energy Science (FES) and Advanced Scientific Computing Research (ASCR).

Implementation of all new businesses focused on automation and giving us the latest. We have the Top Automation News shortlisted that you need to read to remain part of the Tech world. The currently existing systems for the people who are unable to speak can only read their eye or hand movements to decipher what they want to say. But no system is yet invented that can truly give a voice to them. This has changed as three research teams at Columbia University have made a system that collects data from neurons and converts this data into speech.

The world is in the midst of an energy transition. This massive shift aims to move away from reliance on fuels that are destructive to the climate, the environment, and people's well-being. The goal established by the UN is to "ensure access to affordable, reliable, sustainable and modern energy for all" by 2030. While governments, energy companies, and activists dominate the headlines, the progress with infrastructure and technology won't be sufficient. A successful energy transition for the good of all humanity depends on the action of individuals.