Crowd counting is a regression task that estimates the number of people in a scene image, which plays a vital role in a range of safety-critical applications, such as video surveillance, traffic monitoring and flow control. In this paper, we investigate the vulnerability of deep learning based crowd counting models to backdoor attacks, a major security threat to deep learning. A backdoor attack implants a backdoor trigger into a target model via data poisoning so as to control the model's predictions at test time. Different from image classification models on which most of existing backdoor attacks have been developed and tested, crowd counting models are regression models that output multi-dimensional density maps, thus requiring different techniques to manipulate. In this paper, we propose two novel Density Manipulation Backdoor Attacks (DMBA$^{-}$ and DMBA$^{+}$) to attack the model to produce arbitrarily large or small density estimations. Experimental results demonstrate the effectiveness of our DMBA attacks on five classic crowd counting models and four types of datasets. We also provide an in-depth analysis of the unique challenges of backdooring crowd counting models and reveal two key elements of effective attacks: 1) full and dense triggers and 2) manipulation of the ground truth counts or density maps. Our work could help evaluate the vulnerability of crowd counting models to potential backdoor attacks.

Reservoir computers (RCs), also known as echo state networks, are specialized, artificial neural networks. They are powerful and train very fast. However, today RC is not as commonly employed by the machine learning community at large as other classes of neural network models. Unlike most neural networks, the majority of RC weights are not optimized via the backpropagation algorithm. Instead, the weights are generated via a stochastic process that is very sensitive to a few numbers, typically less than 10, called hyper-parameters (HPs). In a simple feed forward neural network, rather than being optimized during training, HPs govern the learning process. The optimization of HPs is a significant challenge to the widespread adoption of the RC. Other classes of neural networks, which are extremely popular today, have faced similar significant challenges in the past. At times, pessimism about neural networks and, thus, AI more broadly, has led to aversion to these models by the scientific community at large, resulting in an AI winter: a "period following a massive wave of hype for AI characterized by a disillusionment that causes a freeze in funding and publications"

The increasing deployment of end use power resources in distribution systems created active distribution systems. Uncontrolled active distribution systems exhibit wide variations of voltage and loading throughout the day as some of these resources operate under max power tracking control of highly variable wind and solar irradiation while others exhibit random variations and/or dependency on weather conditions. It is necessary to control the system to provide power reliably and securely under normal voltages and frequency. Classical optimization approaches to control the system towards this goal suffer from the dimensionality of the problem and the need for a global optimization approach to coordinate a huge number of small resources. Artificial Intelligence (AI) methods offer an alternative that can provide a practical approach to this problem. We suggest that neural networks with self-attention mechanisms have the potential to aid in the optimization of the system. In this paper, we present this approach and provide promising preliminary results.

Recent works have shown how the reasoning capabilities of Large Language Models (LLMs) can be applied to domains beyond natural language processing, such as planning and interaction for robots. These embodied problems require an agent to understand many semantic aspects of the world: the repertoire of skills available, how these skills influence the world, and how changes to the world map back to the language. LLMs planning in embodied environments need to consider not just what skills to do, but also how and when to do them - answers that change over time in response to the agent's own choices. In this work, we investigate to what extent LLMs used in such embodied contexts can reason over sources of feedback provided through natural language, without any additional training. We propose that by leveraging environment feedback, LLMs are able to form an inner monologue that allows them to more richly process and plan in robotic control scenarios. We investigate a variety of sources of feedback, such as success detection, scene description, and human interaction. We find that closed-loop language feedback significantly improves high-level instruction completion on three domains, including simulated and real table top rearrangement tasks and long-horizon mobile manipulation tasks in a kitchen environment in the real world.

Security-constrained unit commitment (SCUC) is solved for power system day-ahead generation scheduling, which is a large-scale mixed-integer linear programming problem and is very computationally intensive. Model reduction of SCUC may bring significant time savings. In this work, a novel approach is proposed to effectively utilize machine learning (ML) to reduce the problem size of SCUC. An ML model using logistic regression (LR) algorithm is proposed and trained with historical nodal demand profiles and the respective commitment schedules. The ML outputs are processed and analyzed to reduce variables and constraints in SCUC. The proposed approach is validated on several standard test systems including IEEE 24-bus system, IEEE 73-bus system, IEEE 118-bus system, synthetic South Carolina 500-bus system and Polish 2383-bus system. Simulation results demonstrate that the use of the prediction from the proposed LR model in SCUC model reduction can substantially reduce the computing time while maintaining solution quality.

We study a human-robot collaborative transportation task in presence of obstacles. The task for each agent is to carry a rigid object to a common target position, while safely avoiding obstacles and satisfying the compliance and actuation constraints of the other agent. Human and robot do not share the local view of the environment. The human policy either assists the robot when they deem the robot actions safe based on their perception of the environment, or actively leads the task. Using estimated human inputs, the robot plans a trajectory for the transported object by solving a constrained finite time optimal control problem. Sensors on the robot measure the inputs applied by the human. The robot then appropriately applies a weighted combination of the human's applied and its own planned inputs, where the weights are chosen based on the robot's trust value on its estimates of the human's inputs. This allows for a dynamic leader-follower role adaptation of the robot throughout the task. Furthermore, under a low value of trust, if the robot approaches any obstacle potentially unknown to the human, it triggers a safe stopping policy, maintaining safety of the system and signaling a required change in the human's intent. With experimental results, we demonstrate the efficacy of the proposed approach.

Sustainable business is a strategy that incorporates environmental, social and governance factors into decision-making, and it is becoming an increasingly important component of business strategy. In fact, in a recent Gartner survey, chief executives identified environmental sustainability as a top 10 business priority for the first time in a decade. Technology is an essential part of the framework that business leaders need to deliver on sustainable business outcomes. However, technology can be a double-edged sword when it comes to sustainability. It can support sustainability goals by improving the quality, scale and impact of environmental initiatives.

The growth of computationally intensive technologies such as machine learning incurs a high carbon footprint and is contributing to climate change. Alongside that rapid growth is an expanding portfolio of green AI tools and techniques to help offset carbon usage and provide a more sustainable path forward. The cost to the environment is high, according to research published last month by Microsoft and the Allen Institute for AI, with co-authors from Hebrew University, Carnegie Mellon University and Hugging Face, an AI community. The study extrapolated data to show that one training instance for a single 6 billion parameter transformer ML model -- a large language model -- is the CO2 equivalent to burning all the coal in a large railroad car, according to Will Buchanan, product manager for Azure machine learning at Microsoft, Green Software Foundation member and co-author of the study. In the past, code was optimized in embedded systems that are constrained by limited resources such as those seen in phones, refrigerators or satellites, said Abhijit Sunil, analyst at Forrester Research.

This piece provides an overview of the current status of quantum computing for those just starting to look at the field. For those interested in learning more, we recommend viewing the video of a recent panel session from the recent HPE Discover 2022 event in Las Vegas, Nevada. Besides myself, other members of the panel included Kirk Bresniker, HPE Fellow, VP and Chief Architect, Hewlett Packard Labs, Yehuda Naveh, co-founder and CTO of Classiq, and Dr. Shini Somara, moderator and TV technology journalist. So, is quantum computing ready to take off and disrupt industries as we know them? As an analyst and publisher about all things quantum, I hear variations of this question every day. My response is to fall back on the decades-old "Amara's Law," which states that: "We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run."

Continuous Normalizing Flows (CNFs) are a class of generative models that transform a prior distribution to a model distribution by solving an ordinary differential equation (ODE). We propose to train CNFs on manifolds by minimizing probability path divergence (PPD), a novel family of divergences between the probability density path generated by the CNF and a target probability density path. PPD is formulated using a logarithmic mass conservation formula which is a linear first order partial differential equation relating the log target probabilities and the CNF's defining vector field. PPD has several key benefits over existing methods: it sidesteps the need to solve an ODE per iteration, readily applies to manifold data, scales to high dimensions, and is compatible with a large family of target paths interpolating pure noise and data in finite time. Theoretically, PPD is shown to bound classical probability divergences. Empirically, we show that CNFs learned by minimizing PPD achieve state-of-the-art results in likelihoods and sample quality on existing low-dimensional manifold benchmarks, and is the first example of a generative model to scale to moderately high dimensional manifolds.