Science-Videos
11.6K views | +0 today
Science-Videos
Learn Science At Home Through Videos
Your new post is loading...
Your new post is loading...
Scooped by Dr. Stefan Gruenwald
Scoop.it!

A Deep Learning Framework For Visual Scene Understanding

Presented at TTI/Vanguard's Networks, Sensors, & Mobility May 3–4, 2016 San Francisco, CA Alex Kendall, Department of Engineering, University of Cambridge

 

We can now teach machines to recognize objects. However, in order to teach a machine to “see” we need to understand geometry as well as semantics. Given an image of a road scene, for example, an autonomous vehicle needs to determine where it is, what's around it, and what's going to happen next. This requires not only object recognition, but depth, motion and spatial perception, and instance-level identification. A deep learning architecture can achieve all these tasks at once, even when given a single monocular input image. Surprisingly, jointly learning these different tasks results in superior performance, because it causes the deep network to uncover a better deep representation by explicitly supervising more information about the scene. This method outperforms other approaches on a number of benchmark datasets, such as SUN RGB-D indoor scene understanding and CityScapes road scene understanding. Besides cars, potential applications include factory robotics and systems to help the blind.

more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Technoscience and the Future
Scoop.it!

Intelligent Autonomous Systems

This talk describes the current research path towards intelligent, semi-autonomous systems, where both humans and automation tightly interact, and together, accomplish tasks such as searching for survivors of a hurricane using a team of UAVs with sensors with highly efficient interaction. This talk is describes the current state of the art in 1) intelligent robotic (only) systems, 2) modeling human decisions and 3) semi-autonomous systems, with a focus on information exchange, and command and control.

 

Mark Campbell is the S.C. Thomas Sze Director of the Sibley School of Mechanical and Aerospace Engineering at Cornell University.


Via Szabolcs Kósa, olsen jay nelson
more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

Videos of machine learning, artificial intelligence and playful machines

Videos of machine learning, artificial intelligence and playful machines | Science-Videos | Scoop.it
more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

Google Workshop on Quantum Biology: Classical and Quantum Information in DNA

DNA stores and replicates information. Special sequences of different nucleic acids (adenine, cytosine, guanine, thymine) encode life's blueprints. These nucleic acids can be divided into a classical part (massive core) and a quantum part (electron shell and single protons). The laws of quantum mechanics map the classical information (A,C,G,T) onto the configuration of electrons and position of single protons. Although DNA replication requires perfect copies of the classical information, the core that constitutes this information does not directly interact with the copying machine. Instead, only the quantum degrees of freedom are measured. Thus successful copying requires a correct translation of classical to quantum to classical information. It has been shown that the electronic system is well shielded from thermal noise. This leads to entanglement inside the DNA helix. It is an open question if this entanglement influences the genetic information processing. In this talk I will discuss possible consequences of entanglement for the information flow and the similarities and differences between classical computing, quantum computing and DNA information processing.

more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

Mandelbulb 3D fractals - Large collection of three dimensional fractal worlds [Videos]

Mandelbulb 3D fractals - Large collection of three dimensional fractal worlds [Videos] | Science-Videos | Scoop.it

3D pictures: http://www.skytopia.com/project/fractal/gallery/

more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

Cognitive Engineering - Engineering Memories - The Future is Now

Dr. Theodore Berger's research is currently focused primarily on the hippocampus, a neural system essential for learning and memory functions.


Theodore Berger leads a multi-disciplinary collaboration with Drs. Marmarelis, Song, Granacki, Heck, and Liu at the University of Southern California, Dr. Cheung at City University of Hong Kong, Drs. Hampson and Deadwyler at Wake Forest University, and Dr. Gerhardt at the University of Kentucky, that is developing a microchip-based neural prosthesis for the hippocampus, a region of the brain responsible for long-term memory. Damage to the hippocampus is frequently associated with epilepsy, stroke, and dementia (Alzheimer's disease), and is considered to underlie the memory deficits characteristic of these neurological conditions.


The essential goals of Dr. Berger's multi-laboratory effort include: (1) experimental study of neuron and neural network function during memory formation -- how does the hippocampus encode information?, (2) formulation of biologically realistic models of neural system dynamics -- can that encoding process be described mathematically to realize a predictive model of how the hippocampus responds to any event?, (3) microchip implementation of neural system models -- can the mathematical model be realized as a set of electronic circuits to achieve parallel processing, rapid computational speed, and miniaturization?, and (4) creation of conformal neuron-electrode interfaces -- can cytoarchitectonic-appropriate multi-electrode arrays be created to optimize bi-directional communication with the brain? By integrating solutions to these component problems, the team is realizing a biomimetic model of hippocampal nonlinear dynamics that can perform the same function as part of the hippocampus.

more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

How to create a Connectome Observatory of the mouse brain and beyond

How to create a Connectome Observatory of the mouse brain and beyond | Science-Videos | Scoop.it

Several laboratories are now using Focused Ion Beam Scanning Electron Microscopes (FIB-SEM) to image small volumes of plastic embedded brain tissue at resolutions approaching 5x5x5nm voxel size. The fact that FIBSEM can obtain such resolution is of fundamental importance since at this resolution all neuronal processes should be traceable with 100% accuracy using fully automatic algorithms. A fundamental physical limitation of the FIB ablation process is that this resolution can only be obtained for very small samples on the order of 20 microns across. To overcome this limitation Ken Hayworth has developed a technique using a heated, oil-lubricated, ultrasonically vibrating diamond knife which can section large blocks of plastic-embedded brain tissue into 20 micron thick strips optimally sized for high-resolution FIB-SEM imaging. Crucially, this thick sectioning procedure results in such high-quality surfaces that the finest neuronal processes can be traced from strip to strip.

more...
No comment yet.
Scooped by Dr. Stefan Gruenwald
Scoop.it!

Self-Improving Artificial Intelligence and the Future of Computing

Steve Omohundro for the Stanford University Computer Systems Colloquium (EE 380) presents fundamental principles that underlie the operation of "self-improving systems," i.e., computer software and hardware that improve themselves by learning from their own operations.

more...
No comment yet.
Scooped by Dr. Stefan Gruenwald
Scoop.it!

NIPS Workshop on Optimization for Machine Learning, Whistler 2008 - Video Lectures

Classical optimization techniques have found widespread use in machine learning. Convex optimization has occupied the center-stage and significant effort continues to be still devoted to it.

 

Pattern Analysis, Statistical Modelling and Computational Learning » NIPS Workshop on Optimization for Machine Learning, Whistler 2008.

 

Training a Binary Classifier with the Quantum Adiabatic Algorithm

 

Polyhedral Approximations in Convex Optimization

 

Optimization in Machine Learning: Recent Developments and Current Challenges

 

Large-scale Machine Learning and Stochastic Algorithms

 

Online and Batch Learning Using Forward-Looking Subgradients

 

Robustness and Regularization of Support Vector Machines

 

Tuning Optimizers for Time-Constrained Problems using Reinforcement Learning.

more...
No comment yet.
Rescooped by Dr. Stefan Gruenwald from Amazing Science
Scoop.it!

Dr. Eric Horvitz and Dr. Peter Norvig: The Challenge and Promise of Artificial Intelligence

Join leading researchers Dr. Eric Horvitz of Microsoft Research and Dr. Peter Norvig of Google for an intriguing discussion about the past, present, and future of artificial intelligence, moderated by KQED's Tim Olson.


Via Szabolcs Kósa, Dr. Stefan Gruenwald
more...
No comment yet.
Scooped by Dr. Stefan Gruenwald
Scoop.it!

Artificial Intelligence - 36 hours of VIDEO lectures

Artificial Intelligence at Ravensburg-Weingarten University - Essential for Master of Computer Science

more...
No comment yet.