Object-oriented representations in reinforcement learning have shown promise in transfer learning, with previous research introducing a propositional objectoriented framework that has provably efficient learning bounds with respect to samplecomplexity.

We introduce point affiliation into feature upsampling, a notion that describes the affiliation of each upsampled point to asemantic cluster formed by local decoder feature points with semantic similarity. By rethinking point affiliation, we present a generic formulation for generating upsampling kernels. The kernels encourage notonly semantic smoothness butalsoboundary sharpness intheupsampled feature maps. Such properties are particularly useful for some dense prediction tasks such as semantic segmentation. The key idea of our formulation istogenerate similarity-awarekernels bycomparing thesimilarity between each encoder feature point and the spatially associated local region of decoder features.

Jerome Pesenti leads the development of artificial intelligence at one of the world's most influential--and controversial--companies. As VP of artificial intelligence at Facebook, he oversees hundreds of scientists and engineers whose work shapes the company's direction and its impact on the wider world. AI is fundamentally important to Facebook. Algorithms that learn to grab and hold our attention help make the platform and its sister products, Instagram and WhatsApp, stickier and more addictive. And, despite some notable AI flops, like the personal assistant M, Facebook continues to use AI to build new features and products, from Instagram filters to augmented reality apps. Mark Zuckerberg has promised to deploy AI to help solve some of the company's biggest problems, by policing hate speech, fake news, and cyberbullying (an effort that has seen limited success so far).

JEEVES's success depended crucially on JEEVES's visual range was extremely JEEVES as successful as it was? JEEVES's success was that its software JEEVES's hardware was designed and built by JEEVES can reverse the direction of the brush. It is equipped with seven ultrasonic proximity sensors (only five were used in the competition), a wide-angle color camera, and a high-speed colorbased vision system manufactured by Newton Research Labs. Prior to the competition, the vision system was trained to recognize yellow tennis balls, pink squiggle balls, and cyan markers that marked the gate. The vision system proved extremely reliable during the competition, benefiting from clear color cues provided by the objects.

This let Flakey drive along hallways with no dead reckoning or planning whatsoever. It seemed miraculous at the time; a situated automaton that knew things without needing any models. However, I thought of it as (sensor-driven) feedback control, versus (plan driven, eyes shut) feed-forward control. I then used Mike Georgeff's procedural reasoning system (PRS) to make Flakey not only drive but navigate an office building. In some respects this project succeeded: the robot's "domain knowledge" was nothing more than a static connection graph--no distances to drive, no widths of halls or doorways, no a priori obstacles--such information was acquired en route from sensory input.

The capabilities for representing and reasoning about three-dimensional (3-D) objects are essential for knowledgebased, 3-D photointerpretation systems that combine domain knowledge with image processing, as demonstrated by 3-D Mosaic and ACRONYM. Three-dimensional representation of objects is necessary for many additional applications, such as robot navigation and 3-D change detection. Geometric reasoning is especially important because geometric relationships between object parts are a rich source of domain knowledge. A practical framework for geometric representation and reasoning must incorporate projections between a two-dimensional (2-D) image and a 3-D scene, shape and surface properties of objects, and geometric and topological relationships between objects. In addition, it should allow easy modification and extension of the system's domain knowledge and be flexible enough to organize its reasoning efficiently to take advantage of the current available knowledge.

It's 2017 so of course the phrase, "a robot cut through a wall to pass the Olympic torch on to its creator" is a real sentence to describe a real thing that actually happened. On Monday, "Hubo," a robot created at the Korean Advanced Institute of Science and Technology, passed the Olympic flame through a hole it cut in a wall to its creator, Professor Oh Jun-Ho. The passing of the torch was a part of the 101-day Olympic torch relay as the 2018 Winter Olympics fast approaches. The hole-cutting party trick was the same task Hubo performed in 2015 to win the DARPA Robotics Challenge. So a congratulations is in order to "Hubo" for carrying the Olympic flame and also to humanity for having handed fire to a robot and lived to tell the tale.

A smart home with walls capable of monitoring each person's health without wearable fitness trackers just got a step closer. A research group based primarily at MIT has shown how a picture-size wall sensor can detect the walking speed and stride length of people based on how their bodies interfere with radio signals. The WiGait system works by transmitting low-power radio signals and analyzing how those radio signals reflect off of people's bodies within a radius of 9 to 12 meters. It can also detect the walking patterns of multiple people and works across physical barriers such as walls within a home. According to its inventors, this "invisible" sensor could combined with wearable devices used to track other fitness measures--especially for older people with chronic conditions such as Parkinson's disease and multiple sclerosis.

The day is now here when people can gather inside a conference room for a meeting and not have to worry when someone asks a question to which no one knows the answer. All someone will have to do is ask the room itself, because it will dig up the answer for everybody within a few seconds. The adage "if these walls could speak," is coming true in the most literal sense. This is exactly what IBM Watson and several of its partners have done with its new Voice-Enabled Cognitive Rooms, starting with the first use case at Thomas Jefferson University Hospital in Philadelphia that was announced April 19. Patients there now interact with in-room speakers that are connected to the IBM Watson IoT Platform and can easily control their in-room subsystems (windows, lights, temperature settings) without having to physically perform any manual tasks or try to figure out new systems. For conference rooms, IBM's cognitive intelligence apps are embedded into Harman microphone soundbars and alarm clocks with other form factors being developed, so that users can interact by using natural language.

It seemed miraculous at the time; a situated automaton that knew things without needing any models. However, I thought of it as (sensor-driven) feedback control, versus (plan driven, eyes shut) feed-forward control. I then used Mike Georgeff's procedural reasoning system (PRS) to make Flakey not only drive but navigate an office building. In some respects this project succeeded: the robot's "domain knowledge" was nothing more than a static connection graph--no distances to drive, no widths of halls or doorways, no a priori obstacles--such information was acquired en route from sensory input. In other respects, however, progress was unsatisfying.