We propose a control design method for linear time-invariant systems that iteratively learns to satisfy unknown polyhedral state constraints. At each iteration of a repetitive task, the method constructs an estimate of the unknown environment constraints using collected closed-loop trajectory data. This estimated constraint set is improved iteratively upon collection of additional data. An MPC controller is then designed to robustly satisfy the estimated constraint set. This paper presents the details of the proposed approach, and provides robust and probabilistic guarantees of constraint satisfaction as a function of the number of executed task iterations. We demonstrate the safety of the proposed framework and explore the safety vs. performance trade-off in a detailed numerical example.

A low-dimensional dynamical system is observed in an experiment as a high-dimensional signal; For example, a video of a chaotic pendulums system. Assuming that we know the dynamical model up to some unknown parameters, can we estimate the underlying system's parameters by measuring its time-evolution only once? The key information for performing this estimation lies in the temporal inter-dependencies between the signal and the model. We propose a kernel-based score to compare these dependencies. Our score generalizes a maximum likelihood estimator for a linear model to a general nonlinear setting in an unknown feature space. We estimate the system's underlying parameters by maximizing the proposed score. We demonstrate the accuracy and efficiency of the method using two chaotic dynamical systems - the double pendulum and the Lorenz '63 model.

A closed-form solution exists in two-class linear discriminant analysis (LDA), which discriminates two Gaussian-distributed classes in a multi-dimensional feature space. In this work, we interpret the multilayer perceptron (MLP) as a generalization of a two-class LDA system so that it can handle an input composed by multiple Gaussian modalities belonging to multiple classes. Besides input layer $l_{in}$ and output layer $l_{out}$, the MLP of interest consists of two intermediate layers, $l_1$ and $l_2$. We propose a feedforward design that has three stages: 1) from $l_{in}$ to $l_1$: half-space partitionings accomplished by multiple parallel LDAs, 2) from $l_1$ to $l_2$: subspace isolation where one Gaussian modality is represented by one neuron, 3) from $l_2$ to $l_{out}$: class-wise subspace mergence, where each Gaussian modality is connected to its target class. Through this process, we present an automatic MLP design that can specify the network architecture (i.e., the layer number and the neuron number at a layer) and all filter weights in a feedforward one-pass fashion. This design can be generalized to an arbitrary distribution by leveraging the Gaussian mixture model (GMM). Experiments are conducted to compare the performance of the traditional backpropagation-based MLP (BP-MLP) and the new feedforward MLP (FF-MLP).

We explore the use of multiple deep learning models for detecting flaws in software programs. Current, standard approaches for flaw detection rely on a single representation of a software program (e.g., source code or a program binary). We illustrate that, by using techniques from multimodal deep learning, we can simultaneously leverage multiple representations of software programs to improve flaw detection over single representation analyses. Specifically, we adapt three deep learning models from the multimodal learning literature for use in flaw detection and demonstrate how these models outperform traditional deep learning models.

State-of-the-art reinforcement learning algorithms mostly rely on being allowed to directly interact with their environment to collect millions of observations. This makes it hard to transfer their success to industrial control problems, where simulations are often very costly or do not exist, and exploring in the real environment can potentially lead to catastrophic events. Recently developed, model-free, offline algorithms, can learn from a single dataset by mitigating extrapolation error in value functions. However, the robustness of the training process is still comparatively low, a problem known from methods using value functions. To improve robustness and stability of the learning process, we use dynamics models to assess policy performance instead of value functions, resulting in MOOSE (MOdel-based Offline policy Search with Ensembles), an algorithm which ensures low model bias by keeping the policy within the support of the data. We compare MOOSE with state-of-the-art model-free, offline RL algorithms BEAR and BCQ on the Industrial Benchmark and Mujoco continuous control tasks in terms of robust performance, and find that MOOSE outperforms its model-free counterparts in almost all considered cases, often even by far.

To help make the process faster and cheaper, Ford recently sought the help of a four-legged robot dog made by Boston Dynamics, a subsidiary of Japanese conglomerate SoftBank Group Corp. "It's a huge breakthrough for us," said Mark Goderis, digital engineering manager at Ford. The robot, nicknamed "Fluffy" by one of Ford's digital engineers, weighs 70 pounds and is equipped with five cameras that give it 360-degree vision, letting it observe what's in front of it and avoid obstacles. It can climb stairs and stabilize itself on slippery surfaces and metal grates using optimization algorithms. It can also access hard-to-reach areas within the plant, as long as they are at least 2 feet wide. The robot dog, officially called "Spot" by Boston Dynamics, costs $74,500.

To help make the process faster and cheaper, Ford recently sought the help of a four-legged robot dog made by Boston Dynamics, a subsidiary of Japanese conglomerate SoftBank Group Corp. "It's a huge breakthrough for us," said Mark Goderis, digital engineering manager at Ford. The robot, nicknamed "Fluffy" by one of Ford's digital engineers, weighs 70 pounds and is equipped with five cameras that give it 360-degree vision, letting it observe what's in front of it and avoid obstacles. It can climb stairs and stabilize itself on slippery surfaces and metal grates using optimization algorithms. It can also access hard-to-reach areas within the plant, as long as they are at least 2 feet wide. The robot dog, officially called "Spot" by Boston Dynamics, costs $74,500.

Ouster, a San Francisco-based lidar startup that launched out of stealth in December 2017, today secured $42 million in funding, bringing its total raised to $140 million. Cofounder and CEO Angus Pacala says the fresh capital will be used to fund product development and support sales internationally. Lidar sensors are at the core of autonomous vehicle systems like those from Waymo, Uber, Aurora, and Cruise. These sensors measure the distance to objects by illuminating them with light and measuring the reflected pulses, and their use cases extend beyond the automotive sector. Lidar sensors are often tapped for obstacle detection and mapping in mining robots, atmospheric studies for space, forestry management, wind farm optimization, speed limit enforcement, and even video games.

This page is a compilation of common mistakes made by bash users. Each example is flawed in some way. Yes, it would be great if you could just treat the output of ls or find as a list of filenames and iterate over it. This entire approach is fatally flawed, and there is no trick that can make it work. You must use an entirely different approach. If a filename contains whitespace, it undergoes WordSplitting. Assuming we have a file named 01 - Don't Eat the Yellow Snow.mp3 in the current directory, the for loop will iterate over each word in the resulting file name: 01, -, Don't, Eat, etc. If a filename contains glob characters, it undergoes filename expansion ("globbing"). If ls produces any output containing a * character, the word containing it will become recognized as a pattern and substituted with a list of all filenames that match it. If the command substitution returns multiple filenames, there is no way to tell where the first one ends and the second one begins. Pathnames may contain any character except NUL. Depending on which platform you're on, which arguments you used (or didn't use), and whether its standard output is pointing to a terminal or not, ls may randomly decide to replace certain characters in a filename with "?", or simply not print them at all. Never try to parse the output of ls. It's an external command whose output is intended specifically to be read by a human, not parsed by a script. That may seem desirable since ls adds a newline, but if the last filename in the list ends with a newline, ... or $() will remove that one also. In the ls examples, if the first filename starts with a hyphen, it may lead to pitfall #3. This causes the entire output of ls to be treated as a single word. Instead of iterating over each file name, the loop will only execute once, assigning to f a string with all the filenames rammed together. Nor can you simply change IFS to a newline.

Artificial Intelligence or AI is a field of Data Science that trains computers to learn from experience, adjust to inputs, and perform tasks of certain cognitive levels. Over the last few years, AI has emerged as a significant data science function and, by utilizing advanced algorithms and computing power, AI is transforming the functional, operational, and strategic landscape of various business domains. AI algorithms are designed to make decisions, often using real-time data. Using sensors, digital data, and even remote inputs, AI algorithms combine information from a variety of different sources, analyze the data instantly, and act on the insights derived from the data. Most AI technologies – from advanced recommendation engines to self-driving cars – rely on diverse deep learning models. By utilizing these complex models, AI professionals are able to train computers to accomplish specific tasks by recognizing patterns in the data. Analytics India Magazine (AIM), in association with Jigsaw Academy, has developed this study on the Artificial Intelligence market to understand the developments of the AI market in India, covering the market in terms of Industry and Company Type. Moreover, the study delves into the market size of the different categories of AI and Analytics startups / boutique firms. As a part of the broad Data Science domain, the Artificial Intelligence technology function has so far been classified as an emerging technology segment. Moreover, the AI market in India has, till now, been dominated by the MNC Technology and the GIC or Captive firms. Domestic firms, Indian startups, and even International Technology startups across various sectors have, so far, not made a significant investment, in terms of operations and scale, in the Indian AI market. Additionally, IT services and Boutique AI & Analytics firms had not, till a couple of years ago, developed full-fledged AI offerings in India for their clients.