AIRLINK 72.59 Increased By ▲ 3.39 (4.9%)
BOP 4.99 Increased By ▲ 0.09 (1.84%)
CNERGY 4.29 Increased By ▲ 0.03 (0.7%)
DFML 31.71 Increased By ▲ 0.46 (1.47%)
DGKC 80.90 Increased By ▲ 3.65 (4.72%)
FCCL 21.42 Increased By ▲ 1.42 (7.1%)
FFBL 35.19 Increased By ▲ 0.19 (0.54%)
FFL 9.33 Increased By ▲ 0.21 (2.3%)
GGL 9.82 Increased By ▲ 0.02 (0.2%)
HBL 112.40 Decreased By ▼ -0.36 (-0.32%)
HUBC 136.50 Increased By ▲ 3.46 (2.6%)
HUMNL 7.14 Increased By ▲ 0.19 (2.73%)
KEL 4.35 Increased By ▲ 0.12 (2.84%)
KOSM 4.35 Increased By ▲ 0.10 (2.35%)
MLCF 37.67 Increased By ▲ 1.07 (2.92%)
OGDC 137.75 Increased By ▲ 4.88 (3.67%)
PAEL 23.41 Increased By ▲ 0.77 (3.4%)
PIAA 24.55 Increased By ▲ 0.35 (1.45%)
PIBTL 6.63 Increased By ▲ 0.17 (2.63%)
PPL 125.05 Increased By ▲ 8.75 (7.52%)
PRL 26.99 Increased By ▲ 1.09 (4.21%)
PTC 13.32 Increased By ▲ 0.24 (1.83%)
SEARL 52.70 Increased By ▲ 0.70 (1.35%)
SNGP 70.80 Increased By ▲ 3.20 (4.73%)
SSGC 10.54 No Change ▼ 0.00 (0%)
TELE 8.33 Increased By ▲ 0.05 (0.6%)
TPLP 10.95 Increased By ▲ 0.15 (1.39%)
TRG 60.60 Increased By ▲ 1.31 (2.21%)
UNITY 25.10 Decreased By ▼ -0.03 (-0.12%)
WTL 1.28 Increased By ▲ 0.01 (0.79%)
BR100 7,546 Increased By 137.4 (1.85%)
BR30 24,809 Increased By 772.4 (3.21%)
KSE100 71,902 Increased By 1235.2 (1.75%)
KSE30 23,595 Increased By 371 (1.6%)
Technology

AI Software offering new doors into your perception

New technology innovations show that scientists can now interpret what a person is able to see, remember, imagine or
Published June 2, 2017

New technology innovations show that scientists can now interpret what a person is able to see, remember, imagine or dream.

In order to decode and predict a person’s imagination, researchers from Kyoto and ATR (Advanced Telecommunications Research) Computational Neuroscience Laboratories University, made use of neural network-based artificial intelligence.

Huffington Post reported Yukiyasu Kamitani, the team leader, stated, “When we gaze at an object, our brains process these patterns hierarchically, starting with the simplest and progressing to more complex features.”

Explaining about the AI software, he said, “The AI we used works on the same principle. Named 'Deep Neural Network', or DNN, it was trained by a group now at Google.”

The results published in Nature Communications showed that the DNN can decode the activity of brain patterns into signal patterns of stimulated neurons when both are shown the similar image.

Apart from that, the researchers also discovered that the higher and lower visual brain areas can better decode respective layers of the DNN which in turn reveals a homology between the neural network and the human brain.

“We tested whether a DNN signal pattern decoded from brain activity can be used to identify seen or imagined objects from arbitrary categories. The decoder takes neural network patterns and compares these with image data from a large database. Sure enough, the decoder could identify target objects with high probability,” described Kamitani.

According to Science Daily, Kamitani further wishes to advance the precision of image identification of their discovery. “Bringing AI research and brain science closer together could open the door to new brain-machine interfaces, perhaps even bringing us closer to understanding consciousness itself,” he concluded.

Copyright Business Recorder, 2017

Comments

Comments are closed.