In this paper, we study the actor-action semantic segmentation problem, which requires joint labeling of both actor and action categories in video frames. One major challenge for this task is that when an actor performs an action, different body parts of the actor provide different types of cues for the action category and may receive inconsistent action labeling when they are labeled independently. To address this issue, we propose an end-to-end region-based actor-action segmentation approach which relies on region masks from an instance segmentation algorithm. Our main novelty is to avoid labeling pixels in a region mask independently - instead we assign a single action label to these pixels to achieve consistent action labeling. When a pixel belongs to multiple region masks, max pooling is applied to resolve labeling conflicts. Our approach uses a two-stream network as the front-end (which learns features capturing both appearance and motion information), and uses two region-based segmentation networks as the back-end (which takes the fused features from the two-stream network as the input and predicts actor-action labeling). Experiments on the A2D dataset demonstrate that both the region-based segmentation strategy and the fused features from the two-stream network contribute to the performance improvements. The proposed approach outperforms the state-of-the-art results by more than 8% in mean class accuracy, and more than 5% in mean class IOU, which validates its effectiveness.

Multi-view datasets are frequently encountered in learning tasks, such as web data mining and multimedia information analysis. Given a multi-view dataset, traditional learning algorithms usually decompose it into several single-view datasets, from each of which a single-view model is learned. In contrast, a multi-view learning algorithm can achieve better performance by cooperative learning on the multi-view data. However, existing multi-view approaches mainly focus on the views that are visible and ignore the hidden information behind the visible views, which usually contains some intrinsic information of the multi-view data, or vice versa. To address this problem, this paper proposes a multi-view fuzzy logic system, which utilizes both the hidden information shared by the multiple visible views and the information of each visible view. Extensive experiments were conducted to validate its effectiveness.

Many relevant tasks require an agent to reach a certain state, or to manipulate objects into a desired configuration. For example, we might want a robot to align and assemble a gear onto an axle or insert and turn a key in a lock. These goal-oriented tasks present a considerable challenge for reinforcement learning, since their natural reward function is sparse and prohibitive amounts of exploration are required to reach the goal and receive some learning signal. Past approaches tackle these problems by exploiting expert demonstrations or by manually designing a task-specific reward shaping function to guide the learning agent. Instead, we propose a method to learn these tasks without requiring any prior knowledge other than obtaining a single state in which the task is achieved. The robot is trained in reverse, gradually learning to reach the goal from a set of start states increasingly far from the goal. Our method automatically generates a curriculum of start states that adapts to the agent's performance, leading to efficient training on goal-oriented tasks. We demonstrate our approach on difficult simulated navigation and fine-grained manipulation problems, not solvable by state-of-the-art reinforcement learning methods.

Activities in reinforcement learning (RL) revolve around learning the Markov decision process (MDP) model, in particular, the following parameters: state values, V; state-action values, Q; and policy, pi. These parameters are commonly implemented as an array. Scaling up the problem means scaling up the size of the array and this will quickly lead to a computational bottleneck. To get around this, the RL problem is commonly formulated to learn a specific task using hand-crafted input features to curb the size of the array. In this report, we discuss an alternative end-to-end Deep Reinforcement Learning (DRL) approach where the DRL attempts to learn general task representations which in our context refers to learning to play the Pong game from a sequence of screen snapshots without game-specific hand-crafted features. We apply artificial neural networks (ANN) to approximate a policy of the RL model. The policy network, via Policy Gradients (PG) method, learns to play the Pong game from a sequence of frames without any extra semantics apart from the pixel information and the score. In contrast to the traditional tabular RL approach where the contents in the array have clear interpretations such as V or Q, the interpretation of knowledge content from the weights of the policy network is more illusive. In this work, we experiment with various Deep ANN architectures i.e., Feed forward ANN (FFNN), Convolution ANN (CNN) and Asynchronous Advantage Actor-Critic (A3C). We also examine the activation of hidden nodes and the weights between the input and the hidden layers, before and after the DRL has successfully learnt to play the Pong game. Insights into the internal learning mechanisms and future research directions are then discussed.

These projects emulate the work that these professionals do throughout industry. Machine learning (ML) is often a project component in all three areas but its use depends on the role. Data Scientists often use ML to uncover insights to drive a business or model users to improve data products. Data Engineers solve engineering challenges to apply ML methods when the amount of data is massive and requires distributed computation. And, as mentioned in our post on "How AI Careers Fit into the Data Landscape", AI focuses on understanding core human abilities and designing algorithms, which often have a ML component, to emulate these processes.

Bruce Newsome reviews the recently published book: "Strategy, Evolution, and War: From Apes to Artificial Intelligence," authored by Kenneth Payne and published by Georgetown University Press. Artificial intelligence (AI) has been explicit in the practices and policies of defence since at least the 1970s, at least in high-capacity countries, given the exponential growth in the power of electronic computing per unit cost. It was already specified in training and forecasting simulations, decision-making aids, targeting aids, robotics, adaptive navigation systems (as in the Tomahawk Cruise Missile), and ballistic missile defence. Any child with a video game could experience AI. AI raced up Western governmental priorities in the 2000s by application to countering terrorism; in 2009, the US escalated its cyber capabilities and authorities, partly on the promise of AI; in 2014, the Russians seemed to know first what the defenders of Ukraine were doing, in part because of integration of AI; and in 2016, Western governments consensually blamed Russia for unprecedented interference in American and other elections, partly aided by AI.

The majority of the deep learning applications that we see in the community are usually geared towards fields like marketing, sales, finance, etc. We hardly ever read articles or find resources about deep learning being used to protect these products, and the business, from malware and hacker attacks. While the big technology companies like Google, Facebook, Microsoft, and Salesforce have already embedded deep learning into their products, the cybersecurity industry is still playing catch up. It's a challenging field but one that needs our full attention. In this article, we briefly introduce Deep Learning (DL) along with a few existing Information Security (hereby referred to as InfoSec) applications it enables. We then deep dive into the interesting problem of anonymous tor traffic detection and also present a DL-based solution to detect TOR traffic.

Does all the hype about AI sound just a little too familiar? If you're old enough to remember the first beginnings of the Internet and the dotcom bubble, you might also remember the tsunami of hype that attended these events as they unfolded. Wired magazine made endlessly breathless predictions about how the Internet would transform humanity and bring about a technologically-driven utopia. Now we're wrestling with how such a promising technology devolved into a netherworld of hacking, hate speech, exploitation of personal data, "dark webs", misinformation, political chicanery, and citizen surveillance despite these glowing promises. In the latest twist, AI is being sold in a similar way by similar players and the cultural amnesia is impressive.

When tasks feel insurmountable, I have always retreated to a tried and true hack, the sort any self-help book worth the price of the Kindle it's living in will dispense: Break the big, scary thing into smaller tasks. The nice news is that, sometimes, the little task ends up being more interesting, more enlightening, more fun, and more doable than the scary, big thing. This week, WIRED Transportation spent some time with the people sweating the small stuff, the tinkerers making adjustments at the peripherals. The German carmakers running a curious mobility experiment in Seattle; the coders making it easier for cities to share the rules of the road with self-driving cars; the engineers coming up with a very special hook that should someday help autonomous drones deliver their wares. Turns out that work is vital, too.

AI is making the headlines these days, from warnings about it being an ingredient to humanity's extinction to praise concerning the new technical possibilities it might bring in the near future. But the current discussion is lacking a realistic view on what AI is and what it is not. Despite this, expectations for the technology are exceptionally high. The reasons for this heightened attention in the public are achievements such as AI beating chess or Go world champions or autonomous cars. This AI breakthrough is a result of the exponential increase in computing power and – even more importantly – data.