Brain-inspired automated visual object discovery and detection

  • UCLA Newsroom | PDF

    UCLA and Stanford University engineers have demonstrated a computer system that can discover and identify the real-world objects it “sees” based on the same method of visual learning that humans use.

  • Photonics | PDF

    The researchers drew insight into contextual learning from findings in cognitive psychology and neuroscience…. 

  • TechXplore | PDF

    …To help the new system "learn" more like humans, the engineers decided to immerse it in an internet replica of the environment humans live in...

  • R&D | PDF

    … It is an important step toward general artificial intelligence systems--computers that learn on their own, are intuitive, make decisions based on reasoning and interact with humans in a more human-like way…

Read the original paper


Interview on Read Before You Cite Dec. 2002. 

“Mommy Blogs” and the Vaccination Exemption Narrative: Results From A Machine-Learning Approach for Story Aggregation on Parenting Social Media Sites

"...[A] multidisciplinary team of UCLA researchers has built an elegant computational model that reflects how humans think and communicate, thereby teaching computers to understand structured narratives within the flow of posts on the internet."

"...[T]he impact could be the creation of a more effective way to help people separate fact from fiction in [...] social media conversations"

"The researchers said their success [...] demonstrates the capability to introduce counter-narratives into internet interactions, break up echo chambers and one day potentially help root out fact from fiction for social media users."

"[The researchers,] using natural language processing methods, were able to identify characters and the relationships between those characters, discovering the core of the underlying narratives"

Read the original paper

Stochastic Modeling of a Serial Killer

"Researchers have discovered that the seemingly erratic behavior of the "Rostov Ripper," a prolific serial killer active in the 1980s, conformed to the same mathematical pattern obeyed by earthquakes, avalanches, stock market crashes and many other sporadic events."

"The implications of the study could give police insight into a killer’s habits and mind from this point forward."

"...[T]here must be a threshold - that is, when a certain number of neurons fire, the serial killer becomes driven by an overwhelming urge to kill."

"The [...] outcome suggests that there was an underlying natural process driving the serial killer's behavior."

"Simkin and Roychowdhury hypothesize that it's the same type of effect that has also been found to cause epileptics to have seizures."

"The research [...] seems to offer the possibility that police could be 'ready' when other killers are driven to strike."

"Researchers [...] found that behavior of murderers may adhere to a pretty strict mathematical formula."

Read the original paper

Physical Fault Tolerance of Nanoelectronics

"Their conclusion—that error prevention is better than error correction—has implications for transistor device technologies and CMOS scaling, and may impose a minimum limit on the size of devices."

"The findings are of immediate relevance to researchers working in transistor-scaling, through to scientists developing new device concepts."

Read the original paper


Stochastic Modeling of Congress

"It turns out that the way a particular resolution gains support can be accurately simulated by the avalanches that occur when grains of sand are dropped onto each other to form a pile." 

Read the original paper

Experience Versus Talent Shapes the Structure of the Web

"The authors found that although experience is an asset, new and talented web sites regularly oust existing leaders."

"Analysing the rise and fall of websites is the perfect way to shed light on the old debate over whether talent or experience matters most, say mathematicians."

"The evolving Internet has been traditionally conservative in spotting talent, showing a preference for experience, which has led to the rich-gets-richer syndrome among websites, but a new study shows that upstart new websites regularly oust existing leaders by providing enhanced novelty and utility."

"Ebenso wichtig wie die "Erfahrung" einer Seite, also die Menge ihrer Verlinkungen, ist das "Talent", also das Potenzial einer Seite, viele Links zu ergattern."

"...[A]s in high school or high-fashion, popularity on the web is a fickle thing..."

Read the orignal article

A Theory of Web Traffic

"Although the Internet contains well over 100 million Web sites, two electrical engineers think they know what the traffic patterns of the entire Web look like."

Read the original article

Capacity Constraints and the Inevitability of Mediators in Adword Auctions                        

"The mainstay of the US $14 billion online-advertising industry—the 'sponsored links' that appear alongside Google and Yahoo search results—contains an inherent inefficiency that is destined to give birth to third-party search-term auctioneers and fuel the growth of so-called mediators"

Read the original paper

Theory of Aces: High Score by Skill or Luck?

"The legend of Manfred von Richthofen, aka the Red Baron, has taken a knock. The victories notched up by him and other great flying aces of the first world war could have been down to luck rather than skill."

"...while the first paper was mildly controversial with Red Baron fans, the new one is tantamount to dropping an insulting note tied to a spanner, right smack dab into the middle of their aerodrome."

"Throughout human history, people have debated whether success is determined by luck or talent. A new analysis of the performance of the Red Baron in World War I offers an interesting addendum to this debate."

Read the original paper

Scalable Percolation Search in Power Law Networks

"Researchers from the University of California at Los Angeles have devised a fast search algorithm that uses local rules to find nodes and content in randomly-formed, scale-free networks such as the Internet"

Read the original paper

Scale-Free and Stable Structures in Complex Ad Hoc Networks

"...schemes that will optimize the organization and sharing of information among dynamic groups of file swappers."

Read the original paper

Personal Email Networks: An Effective Anti-Spam Tool

"A simple and easily implemented scheme for combating e-mail spam... The method effectively turns the spammers' weapon on themselves. "

Read the original paper

Collaborative Spam Filtering Using Email Networks

"The technique exploits the structure of social networks to determine whether a message is from a friend or a spammer."

"More than two-thirds of all email traffic is spam. But the problem could be reduced if our computers work together to control it."

"Combatting spam needs to be a joint effort, a pooling of anti-spam programme information the way police join forces to catch criminals"

Read the original paper 

Copied Citations Create Renowned Papers?

"Simple mathematical probability, not genius, can explain why some papers are cited a lot more than others"

"Scientific papers that are not widely read and that lack any great influence can end up being classed as high-impact"

Read the original paper

Theory of Aces: Fame By Chance or Merit?

"With the newspapers filled with stories about minor celebrities, you might think it is easy to become famous. But a theory of fame developed by statisticians suggests people are far more likely to languish unrecognised while a few become more famous than they deserve."

Read the original paper

Read Before You Cite!

"Many of the references cited in scientific papers have not been read by the authors citing them"

"A cunning statistical study has exposed scientists as sloppy reporters"

"By examining these mutations we can trace these errors backward in time, and understand how knowledge truly spread from scientist to scientist, instead of how it appeared to spread."

"...negative data is inherently less cited and therefore tends to fail to spread through citation networks."

Read the original paper

Chess Players’ Fame Versus Their Merit

"[While] it’s easy to measure fame by the number of Google hits, it’s much harder to measure achievement. So comparisons are hard to make. ...Simkin and Roychowdhury... have found a way to solve this problem using the well-studied achievement of top-ranked chess players."

Read the original paper