Click header name below to compare STT with manual transcript:
manual | watson | voxolab | microsoft | |
00:00:00,030 --> 00:00:03,360 [3.166000] | 00:00:00,030 --> 00:00:03,600 [3.166000] | 00:00:00,260 --> 00:00:03,360 [2.936000] | 00:00:00,000 --> 00:00:03,196 [3.196000] | 00:00:00,020 --> 00:00:03,993 [3.176000] |
Artificial intelligence is said to have huge impacts our own lives and | sure intelligence is said to have huge impacts our own lives and even | intelligence is said to have huge impact on your life and | — Visual intelligence is said to have huge impacts on our lives | Special intelligence is said to have huge impacts our own lives |
00:00:03,360 --> 00:00:07,170 [5.724000] | 00:00:03,600 --> 00:00:09,630 [5.484000] | 00:00:03,360 --> 00:00:07,920 [5.724000] | 00:00:03,196 --> 00:00:09,084 [5.888000] | 00:00:03,993 --> 00:00:08,040 [5.091000] |
even society. And some actually say it's the new | society %HESITATION and some actually say it's the new electricity since | even society and some actually say it's the new electricity | and even society's and some actually say it's the new electricity | and even society and some actually say it's the new electricity. |
00:00:07,920 --> 00:00:10,670 [2.750000] | 00:00:09,630 --> 00:00:13,380 [1.040000] | 00:00:09,330 --> 00:00:13,380 [1.340000] | 00:00:09,084 --> 00:00:12,868 [1.586000] | 00:00:09,380 --> 00:00:14,161 [1.290000] |
electricity. Since the fifties, | the fifties %HESITATION computer sciences have been trying to program | since the 50s Computer Sciences have been trying to program | since the fifties computer scientists have been trying to | Since the 50s of computer scientists have been trying to program |
00:00:10,670 --> 00:00:14,400 [3.730000] | 00:00:13,380 --> 00:00:18,960 [1.020000] | 00:00:13,380 --> 00:00:18,960 [1.020000] | 00:00:12,868 --> 00:00:18,431 [1.532000] | 00:00:14,161 --> 00:00:16,640 [0.239000] |
computer scientists have been trying to program computers to behave | computers to behave like humans so let's say for example you wanted | computers to behave like humans so let's say for example you wanted | program computers to behave like humans so let's say for example | computers to behave like humans. |
00:00:14,840 --> 00:00:17,200 [2.360000] | ||||
like humans. | ||||
00:00:17,200 --> 00:00:20,090 [2.324000] | 00:00:19,020 --> 00:00:24,820 [0.504000] | 00:00:19,020 --> 00:00:24,440 [0.504000] | 00:00:18,431 --> 00:00:24,519 [1.093000] | 00:00:17,190 --> 00:00:19,524 [2.334000] |
So let's say for example you want to teach a computer what a | teacher computer what a cat is well how hard can that be right I mean | to teach your computer what a cat is how hard can that be right | you want to teach a computer what it kept its but can that be right | So let's say for example, when you want to teach |
00:00:20,360 --> 00:00:22,870 [0.600000] | 00:00:19,524 --> 00:00:20,960 [1.436000] | |||
cat is. Well. | a computer whether catches. | |||
00:00:22,870 --> 00:00:26,140 [1.279000] | 00:00:22,370 --> 00:00:24,149 [1.779000] | |||
How hard can that be, right? I mean after all, a cat is just a | Well, how hard can that be right. | |||
00:00:26,140 --> 00:00:30,290 [0.315000] | 00:00:24,820 --> 00:00:28,870 [1.635000] | 00:00:24,550 --> 00:00:27,930 [1.905000] | 00:00:24,519 --> 00:00:28,127 [1.936000] | 00:00:24,149 --> 00:00:26,455 [2.306000] |
collection of shapes and colors. So, computer scientists | after all the cat is just a collection of shapes and colors and does so | I mean after all a cat is just a collection of shapes and colors | i mean after all a cat is just a collection of shapes and colors | I mean after all, a cat is just a collection |
00:00:28,870 --> 00:00:32,140 [0.221000] | 00:00:28,230 --> 00:00:31,460 [0.861000] | 00:00:28,127 --> 00:00:31,809 [0.964000] | 00:00:26,455 --> 00:00:29,091 [2.636000] | |
computers by insists just thought okay we'll just give rules to | and I was so confused I insist just thought okay we'll just | and also computers that insists just thought okay will just give | of shapes and colors and uh so computers thought | |
00:00:29,091 --> 00:00:30,343 [1.252000] | ||||
insists just thought. | ||||
00:00:30,290 --> 00:00:33,320 [3.030000] | 00:00:32,140 --> 00:00:35,880 [1.180000] | 00:00:31,460 --> 00:00:34,651 [1.860000] | 00:00:31,809 --> 00:00:34,906 [1.511000] | 00:00:30,343 --> 00:00:32,715 [2.977000] |
just thought okay we'll just give rules to the computer and it'll be | the computer and she'll be able to manage what ends recognize what | give rules to the computer and she'll be able to manage web | rules of the computer and still be able to manage what s and to | Okay, well, just give rules to the computer |
00:00:33,320 --> 00:00:36,670 [0.910000] | 00:00:32,715 --> 00:00:34,230 [1.515000] | |||
able to manage well, to recognize what a cat is. | and he'll be able to manage. | |||
00:00:36,670 --> 00:00:40,110 [1.522000] | 00:00:35,880 --> 00:00:39,300 [2.312000] | 00:00:34,651 --> 00:00:37,650 [3.541000] | 00:00:34,906 --> 00:00:38,533 [3.286000] | 00:00:34,450 --> 00:00:38,192 [3.742000] |
So, we should just tell the computer, | a cat is so used we should just tell to %HESITATION the computer | chin to recognize what a cat is so you should we should just | recognise what a cat is so used we should just tell it through | What can't recognize what a cat is so you should we should just |
00:00:40,110 --> 00:00:44,220 [1.780000] | 00:00:39,450 --> 00:00:43,230 [2.440000] | 00:00:37,650 --> 00:00:41,890 [4.240000] | 00:00:38,533 --> 00:00:42,051 [3.357000] | 00:00:38,192 --> 00:00:42,081 [3.698000] |
in mathematical language of course, that a cat has a round face, two | in my stomach mathematical language of course that the cat has a round | tell to the computer and Mathematics mathematical language of course | the computer in mathematics mathematical language of course | tell to um the computer in mathematics mathematical language. |
00:00:44,220 --> 00:00:47,860 [1.522000] | 00:00:43,230 --> 00:00:49,400 [2.512000] | 00:00:42,080 --> 00:00:44,600 [3.662000] | 00:00:42,051 --> 00:00:45,742 [3.691000] | 00:00:42,081 --> 00:00:44,649 [3.661000] |
pointy ears, a chubby body, and a long tail. | face two point eight years of chubby body and a long tail but what about | that the cat has a round face 2.8 | that a cat has a round face two point eight years of chevy body | Of course, that the cat has a round face 2.8 |
00:00:48,150 --> 00:00:51,160 [4.280000] | 00:00:49,400 --> 00:00:54,860 [3.030000] | 00:00:44,600 --> 00:00:52,430 [7.830000] | 00:00:45,742 --> 00:00:52,993 [6.688000] | 00:00:44,649 --> 00:00:46,850 [7.781000] |
But. what about this cat? | this cat and what about if the cat is hidden that doesn't | years of chubby body and but what about this cat and what | and a long tail about this kept and what about if the cat is | years of Chubby body and a long tail. |
00:00:51,470 --> 00:00:54,390 [0.883000] | 00:00:47,930 --> 00:00:52,353 [4.423000] | |||
And, what about if the cat is hidden? | But what about this cat an what about if the cat is | |||
00:00:54,390 --> 00:00:57,990 [2.718000] | 00:00:54,860 --> 00:00:59,120 [2.248000] | 00:00:52,430 --> 00:00:57,470 [4.678000] | 00:00:52,993 --> 00:00:58,337 [4.115000] | 00:00:52,353 --> 00:00:57,108 [4.755000] |
That doesn't work. Giving rules to computers will only | work %HESITATION giving will sue computer will only get you so far | about is the cat is hidden that does work giving rules to computer | hidden that doesn't work giving rules to computer will only get | hidden that doesn't work, giving rules to computer |
00:00:57,990 --> 00:01:02,720 [1.330000] | 00:00:57,470 --> 00:01:06,070 [1.850000] | 00:00:58,337 --> 00:01:06,637 [0.983000] | 00:00:57,108 --> 00:00:59,320 [2.212000] | |
get you so far. So another... | will only get you so far so another another type of another | you so far so another another type of this is another type of | will only get you so far. | |
00:01:02,720 --> 00:01:06,400 [3.350000] | 00:01:00,330 --> 00:01:06,070 [5.740000] | 00:01:00,360 --> 00:01:04,390 [5.710000] | ||
another type of... another type of | so another %HESITATION another type of %HESITATION but that is another | So, in other um another typeof. | ||
00:01:06,400 --> 00:01:09,390 [2.601000] | 00:01:06,070 --> 00:01:09,390 [2.931000] | 00:01:06,070 --> 00:01:08,950 [2.931000] | 00:01:06,637 --> 00:01:10,324 [2.364000] | 00:01:05,130 --> 00:01:09,001 [3.871000] |
approach was developed and this new type of approach developed | type of approach was developed and then you type of approach developed | type of approach was developed in the new type of approach | approach was developed and you type of approach development is called | But this is another type of approach was developed in the new type of |
00:01:09,390 --> 00:01:11,430 [3.120000] | 00:01:09,390 --> 00:01:13,010 [3.120000] | 00:01:08,950 --> 00:01:12,510 [3.560000] | 00:01:10,324 --> 00:01:13,996 [2.186000] | 00:01:09,001 --> 00:01:13,080 [3.509000] |
was called machine learning. | where it is called machine learning machine learning gets computers | developed is called machine learning machine learning kids | machine learning machine learning gets computers the ability to | approach developed with is called machine learning machine learning |
00:01:11,430 --> 00:01:15,270 [3.840000] | 00:01:13,010 --> 00:01:17,770 [2.260000] | 00:01:12,510 --> 00:01:16,660 [2.760000] | 00:01:13,996 --> 00:01:18,090 [1.274000] | 00:01:13,080 --> 00:01:17,435 [2.190000] |
Machine learning gives computers the ability to learn, without being | the ability to learn without being explicitly programs and so rather | computers the ability to learn without being explicitly programmed | learn without being explicitly programmed so rather than hand | gives computers the ability to learn without being explicitly programmed |
00:01:15,270 --> 00:01:18,710 [3.440000] | 00:01:17,770 --> 00:01:22,270 [0.940000] | 00:01:17,070 --> 00:01:20,670 [1.640000] | 00:01:18,090 --> 00:01:23,279 [0.620000] | 00:01:17,435 --> 00:01:20,960 [1.275000] |
explicitly programmed. And so, rather than hand coding | than hand coding specific sets of instructions the machinist trained | at so rather than hand coding specific sets of instructions | coding specific sets of instructions the machine is trained using large | as so rather than hand coding specific sets of instructions. |
00:01:18,710 --> 00:01:23,500 [4.790000] | 00:01:22,470 --> 00:01:27,760 [1.030000] | 00:01:20,940 --> 00:01:25,740 [2.560000] | 00:01:23,279 --> 00:01:29,037 [0.221000] | 00:01:21,010 --> 00:01:25,608 [2.490000] |
specific sets of instructions, the machine is trained, using large | using large amounts of data about five years ago %HESITATION a specific | the machine is trained using large amounts of data about 5 | amounts of data about five years ago a specific type of algorithm | The machine is trained using large amounts of data about 5 years ago, |
00:01:23,500 --> 00:01:27,170 [3.670000] | 00:01:25,740 --> 00:01:30,140 [1.430000] | 00:01:25,608 --> 00:01:29,723 [1.562000] | ||
amounts of data. About five years ago, | years ago a specific type of algorithm west of machine learning | a specific type of algorithm was machine learning algorithm | ||
00:01:27,170 --> 00:01:30,950 [3.780000] | 00:01:27,760 --> 00:01:31,560 [3.190000] | 00:01:30,140 --> 00:01:33,310 [0.810000] | 00:01:29,037 --> 00:01:32,702 [1.913000] | 00:01:29,723 --> 00:01:33,353 [1.227000] |
a specific type of algorithm was, of machine learning algorithm was | type of algorithm was of machine learning algorithm was developed | algorithm was developed called Deep learning you might have | was a machine learning algorithm was developed called deep learning | was developed called deep learning you may have heard |
00:01:31,560 --> 00:01:34,010 [2.450000] | 00:01:31,760 --> 00:01:35,150 [2.250000] | 00:01:33,310 --> 00:01:37,910 [0.700000] | 00:01:32,702 --> 00:01:36,465 [1.308000] | 00:01:33,353 --> 00:01:35,370 [0.657000] |
developed, called deep learning. You might have heard of deep | called deep learning you might have heard of deep learning algorithms | heard of deep learning algorithms they were used to beat humans | you might have heard of deep learning algorithms that were | of deep learning algorithms. |
00:01:34,010 --> 00:01:36,990 [2.980000] | 00:01:35,350 --> 00:01:40,180 [1.640000] | 00:01:36,465 --> 00:01:43,342 [0.525000] | 00:01:35,570 --> 00:01:42,090 [1.420000] | |
learning algorithms, that they were used to | that they were used to beat humans playing go or recently | used to beat humans plane go or a recently poker and deep learning | They were used to beat humans playing go or recently poker. | |
00:01:36,990 --> 00:01:42,240 [5.250000] | 00:01:40,180 --> 00:01:45,910 [2.060000] | 00:01:38,090 --> 00:01:44,570 [4.150000] | ||
beat humans playing go, or recently poker. | poker %HESITATION NDP learning uses techniques third let's say loosely | playing go or a recently poker I'm in deep learning uses techniques | ||
00:01:42,240 --> 00:01:45,460 [3.443000] | 00:01:44,570 --> 00:01:49,840 [1.113000] | 00:01:43,342 --> 00:01:47,550 [2.341000] | 00:01:42,230 --> 00:01:45,683 [3.453000] | |
Deep learning uses techniques that are, let's say | that might a Loosely inspired from how our brain works and we're | uses techniques there might say loosely inspired from how our brain | Men deep learning uses techniques start. | |
00:01:45,460 --> 00:01:48,850 [3.390000] | 00:01:45,910 --> 00:01:52,230 [2.940000] | 00:01:47,550 --> 00:01:53,550 [1.300000] | 00:01:45,683 --> 00:01:50,122 [3.167000] | |
loosely inspired from how our brain works. | inspired from how our brain works and where had foreseen very extremely | works and we had verve seeing very extremely where the extremely | Let's say loosely inspired from our brain works and we | |
00:01:49,330 --> 00:01:52,310 [2.980000] | 00:01:49,840 --> 00:01:55,190 [2.470000] | 00:01:50,122 --> 00:01:55,350 [2.188000] | ||
And, we've seen, very extremely, | had worth seeing very extremely extremely impressive results | had were seeing very extremely extremely impressive results. | ||
00:01:52,500 --> 00:01:56,170 [5.050000] | 00:01:52,310 --> 00:01:57,550 [5.240000] | 00:01:56,170 --> 00:02:00,980 [1.380000] | 00:01:53,930 --> 00:01:59,560 [3.620000] | 00:01:56,170 --> 00:02:00,188 [1.380000] |
well, extremely impressive results | well yeah extremely are impressive results because they have a unique | because they have a unique ability to identify patterns from data | impressive results because they have a unique ability to identify | Because they have a unique ability is to identified patterns |
00:01:56,170 --> 00:01:58,560 [2.390000] | 00:01:57,580 --> 00:02:03,620 [0.980000] | |||
because they have, a unique ability | ability is to identify patterns from data and learn from it so this | |||
00:01:58,560 --> 00:02:03,150 [4.590000] | 00:02:00,980 --> 00:02:05,720 [2.170000] | 00:01:59,680 --> 00:02:04,590 [3.470000] | 00:02:00,188 --> 00:02:02,000 [2.962000] | |
to identify patterns from data and learn from it. | and learn from it to this means that these new types of algorithms | patterns from debt and learn from it so this means that these new | from data and learn from it. | |
00:02:03,150 --> 00:02:06,000 [2.850000] | 00:02:03,620 --> 00:02:07,540 [2.380000] | 00:02:04,610 --> 00:02:10,328 [1.390000] | 00:02:03,220 --> 00:02:07,815 [2.780000] | |
So this means that these new types of algorithms | means that these new types of algorithms can %HESITATION can do | types of algorithms can do it let's say three main can solve the main | So this means that these new types of algorithms can um can do it. | |
00:02:06,360 --> 00:02:07,950 [1.590000] | 00:02:07,540 --> 00:02:12,920 [0.410000] | 00:02:07,240 --> 00:02:12,920 [0.710000] | 00:02:07,815 --> 00:02:10,850 [0.135000] | |
can... can do... | it let's say three main can solve three main problems so fed with | can I take three main consult three main problems to fade with | Let's say 3 main consult remain problems. | |
00:02:07,950 --> 00:02:12,020 [4.070000] | 00:02:10,328 --> 00:02:16,409 [1.692000] | |||
let's say three main, can solve three main problems. | problems so faced with are these algorithms can generate new stuff | |||
00:02:12,020 --> 00:02:17,100 [5.080000] | 00:02:12,920 --> 00:02:18,580 [4.180000] | 00:02:12,920 --> 00:02:18,110 [4.180000] | 00:02:16,409 --> 00:02:20,820 [0.691000] | 00:02:12,020 --> 00:02:16,397 [5.080000] |
So, fed with data, these algorithms can generate new stuff. | data these algorithms can generate new stuff so for example if you train | data these organisms can generate new stuff so for example if | so for example if you train a machine a machine learning | So fed with data these algorithms can generate new stuff, |
00:02:17,100 --> 00:02:19,850 [1.486000] | 00:02:18,110 --> 00:02:22,460 [0.476000] | 00:02:16,397 --> 00:02:18,586 [2.189000] | ||
So for example if you train a machine | you train a machine a machine learning algorithm with a set | so for example, if you train. | ||
00:02:19,850 --> 00:02:24,040 [2.750000] | 00:02:18,580 --> 00:02:22,600 [4.020000] | 00:02:22,460 --> 00:02:26,440 [0.140000] | 00:02:20,850 --> 00:02:25,707 [1.750000] | 00:02:18,586 --> 00:02:22,964 [4.014000] |
a machine learning algorithm, with a set of artist painting, | emotional %HESITATION and machine learning algorithm with a set of | of artist painting like Picasso you feed the machine learning | algorithm with a set of artists painting like picasso you feet | A mashing machine learning algorithm with a set of artist |
00:02:24,040 --> 00:02:26,850 [2.990000] | 00:02:22,600 --> 00:02:27,030 [4.430000] | 00:02:26,440 --> 00:02:29,830 [0.590000] | 00:02:25,707 --> 00:02:28,900 [1.323000] | 00:02:22,964 --> 00:02:24,970 [4.066000] |
like Picasso. You feed of machine learning algorithm | artist painting like Picasso feet of machine learning algorithm with | algorithm with Picasso paintings he will be able to reproduce | a machine learning algorithm with picasso paintings he will be able | painting like because so? |
00:02:26,850 --> 00:02:32,020 [1.955000] | 00:02:27,030 --> 00:02:32,520 [1.775000] | 00:02:25,320 --> 00:02:28,805 [3.485000] | ||
with Picasso paintings, he will be able to reproduce, but style, | Picasso paintings he will be able to reproduce but style and I plan | Feed machine learning algorithm with because of paintings, | ||
00:02:30,070 --> 00:02:34,760 [0.990000] | 00:02:28,920 --> 00:02:34,140 [2.140000] | 00:02:28,805 --> 00:02:31,060 [2.255000] | ||
that style in a plant on a new painting and actually we talked | to reproduce that style and a plan on a new painting and actually | he will be able to reproduce that style. | ||
00:02:32,020 --> 00:02:34,980 [2.960000] | 00:02:32,520 --> 00:02:37,060 [2.460000] | 00:02:34,760 --> 00:02:39,150 [0.220000] | 00:02:34,140 --> 00:02:38,232 [0.840000] | 00:02:32,060 --> 00:02:34,674 [2.920000] |
and apply on a new painting, and actually, we talked about | on a new painting and actually we talked about Christmas very briefly | about Christmas very briefly earlier priests my uses machine | we talked about christmas very briefly earlier priest my uses | And a plant on a new painting and actually. |
00:02:35,570 --> 00:02:38,800 [3.286000] | 00:02:37,060 --> 00:02:41,220 [1.796000] | 00:02:38,232 --> 00:02:42,977 [0.624000] | 00:02:34,674 --> 00:02:38,856 [4.182000] | |
Prisma very briefly earlier, Prisma uses | earlier priest my uses machine learning algorithms to do just that | machine learning algorithms to do just that so it can generate | We talked about Prisma very briefly earlier priest my uses machine | |
00:02:38,800 --> 00:02:42,110 [3.310000] | 00:02:39,150 --> 00:02:43,520 [2.960000] | 00:02:38,856 --> 00:02:42,813 [3.254000] | ||
machine learning algorithms to do just that. | learning algorithms to do just that so I can generate you images | learning algorithms to do just that it can generate new images. | ||
00:02:42,110 --> 00:02:45,370 [3.260000] | 00:02:42,110 --> 00:02:45,200 [3.260000] | 00:02:43,520 --> 00:02:46,920 [1.850000] | 00:02:42,977 --> 00:02:46,200 [2.393000] | 00:02:42,813 --> 00:02:45,203 [2.557000] |
So, you can generate new images, you can generate for example new texts, | so you can generate new images it can generate for example new texts | it can generate for example new text it's used to do translations | new images we can generate for example new text it's used to do | It can generate for example, new text. |
00:02:45,370 --> 00:02:51,170 [1.700000] | 00:02:45,370 --> 00:02:52,040 [1.700000] | 00:02:46,200 --> 00:02:52,786 [0.870000] | 00:02:45,203 --> 00:02:47,070 [1.867000] | |
it's used to do translations. It can also do... | its use to do translations %HESITATION it can also do about that that Billy | translations it can also do not that their ability to recognize | It's used to do translations. | |
00:02:47,460 --> 00:02:48,290 [0.830000] | ||||
Um. | ||||
00:02:51,170 --> 00:02:55,000 [2.906000] | 00:02:52,040 --> 00:02:56,010 [2.036000] | 00:02:49,840 --> 00:02:53,700 [4.236000] | 00:02:52,786 --> 00:02:57,237 [1.290000] | 00:02:49,770 --> 00:02:54,076 [4.306000] |
Its ability to recognize patterns is also very useful, for | to recognize pattern is also very useful for predictions to make | it can also do about that ability to recognize pattern is also | pattern is also very useful for predictions to make predictions | It can also do velvet ability to recognize pattern is also very useful |
00:02:55,000 --> 00:02:59,130 [2.857000] | 00:02:56,010 --> 00:02:59,380 [1.847000] | 00:02:53,700 --> 00:02:57,857 [4.157000] | 00:02:57,237 --> 00:03:01,210 [0.620000] | 00:02:54,076 --> 00:02:56,452 [3.781000] |
predictions, to make predictions. I don't know if you've done those types of tests, | predictions %HESITATION I don't if done those type of tests you know | very useful for predictions to make predictions I don't know | i don't if dent those type of test you know cut i q tests or | for predictions to make predictions. |
00:02:59,130 --> 00:03:03,770 [0.588000] | 00:02:59,380 --> 00:03:04,300 [0.338000] | 00:02:57,857 --> 00:03:00,340 [1.861000] | 00:02:56,452 --> 00:02:59,718 [3.266000] | |
you know, kind of IQ tests, or well, this type of tests where they | kind of IQ tests or well this have if this were there and they give | if you've done those type of test you know what kind of IQ | I don't know if you've done those type of test you know | |
00:03:00,340 --> 00:03:06,530 [1.160000] | 00:02:59,718 --> 00:03:01,500 [1.782000] | |||
test or this type of thing they give you a set of a set of | what kind of like you test or. | |||
00:03:03,920 --> 00:03:08,510 [3.340000] | 00:03:04,300 --> 00:03:09,000 [2.960000] | 00:03:06,530 --> 00:03:10,470 [0.730000] | 00:03:02,310 --> 00:03:07,260 [4.950000] | 00:03:02,360 --> 00:03:05,818 [4.900000] |
well, they give you a set of of numbers, or shapes and then | you a set of %HESITATION of numbers or shapes and then you're supposed | numbers or shapes and then you're supposed to find the next one's | well this type of this further they gave you a set of of numbers | Well, this type of things for their they give you a set of um |
00:03:08,510 --> 00:03:11,580 [1.342000] | 00:03:09,000 --> 00:03:12,820 [0.852000] | 00:03:07,410 --> 00:03:11,570 [2.442000] | 00:03:05,818 --> 00:03:09,852 [4.034000] | |
you're supposed to find the next one. | to find the next what's the next one so busy were you doing there | or shapes and then your supposed to find the next's the next one | of numbers or shapes and then you're supposed to find the next ones. | |
00:03:09,852 --> 00:03:10,645 [0.793000] | ||||
The next one. | ||||
00:03:11,580 --> 00:03:14,830 [1.010000] | 00:03:10,660 --> 00:03:13,800 [1.930000] | 00:03:11,570 --> 00:03:14,850 [1.020000] | 00:03:10,645 --> 00:03:12,590 [1.945000] | |
So basically, what you're doing there is identifying patterns, and then | the next one so they see what you're doing there is identifying | so basic what you doing there is identifying patterns and then | So basically what you're doing. | |
00:03:14,830 --> 00:03:18,250 [1.650000] | 00:03:12,820 --> 00:03:16,630 [3.660000] | 00:03:13,950 --> 00:03:17,660 [2.530000] | 00:03:14,870 --> 00:03:18,270 [1.610000] | 00:03:12,590 --> 00:03:16,480 [3.890000] |
predict what's coming next, and computers, and machine learning | is identifying patterns and then predict what's coming next and | patterns and then predict what's coming next and computers and | predict what's coming next and computers and machine learning | There is a defying patterns and then predict what's coming next. |
00:03:18,920 --> 00:03:21,570 [0.687000] | 00:03:16,630 --> 00:03:20,420 [2.977000] | 00:03:17,660 --> 00:03:21,080 [1.947000] | 00:03:18,290 --> 00:03:23,519 [1.317000] | 00:03:16,510 --> 00:03:19,607 [3.097000] |
algorithms, are extremely good at doing that. | computers and machine learning algorithms are extremely good at | machine learning algorithms are extremely good at doing that | algorithms are extremely good at doing that so that's very useful | And computers in machine learning algorithms are |
00:03:20,420 --> 00:03:24,040 [1.399000] | 00:03:21,570 --> 00:03:26,280 [0.249000] | 00:03:19,607 --> 00:03:21,819 [2.212000] | ||
doing that so %HESITATION that's very useful for example in | so that's very useful for example in the business world if you train | extremely good at doing that, so um. | ||
00:03:21,970 --> 00:03:24,040 [1.102000] | 00:03:21,819 --> 00:03:23,072 [1.253000] | |||
So, that's very useful, for example in | That's very useful. | |||
00:03:24,040 --> 00:03:27,390 [1.245000] | 00:03:24,040 --> 00:03:27,390 [1.245000] | 00:03:23,519 --> 00:03:27,380 [1.766000] | 00:03:23,072 --> 00:03:25,285 [2.213000] | |
the business world. If you train an algorithm like that | the business world %HESITATION if you train an algorithm like that | for example in the business world if you train an algorithm like | For example, in the business world. | |
00:03:27,390 --> 00:03:31,720 [1.950000] | 00:03:27,390 --> 00:03:31,720 [1.950000] | 00:03:26,280 --> 00:03:30,570 [3.060000] | 00:03:27,380 --> 00:03:31,871 [1.960000] | 00:03:25,285 --> 00:03:29,340 [4.055000] |
on historical sales data, it can help you forecast the demand | on history called sales that up it can help you forecast the demand | an algorithm like that on historical sales data it can | that on historical sales data it can help you forecast the demand | If you train, an algorithm like that on history called sales data. |
00:03:31,720 --> 00:03:36,770 [0.980000] | 00:03:31,720 --> 00:03:37,460 [0.980000] | 00:03:30,570 --> 00:03:35,670 [2.130000] | 00:03:31,871 --> 00:03:37,899 [0.829000] | 00:03:30,320 --> 00:03:32,700 [2.380000] |
of a product. These algorithms also very good at | of a product these algorithms also very good and may seek making | help you forecast the demand of a product these are great times | of a product these algorithms also very good and music making | It can help you forecast the demand of a product. |
00:03:36,770 --> 00:03:39,810 [1.940000] | 00:03:37,460 --> 00:03:40,790 [1.250000] | 00:03:35,670 --> 00:03:39,220 [3.040000] | 00:03:37,899 --> 00:03:42,743 [0.811000] | 00:03:34,820 --> 00:03:38,710 [3.890000] |
making associations. So I'm going to share with you an | associations so I'm going to share with you an example but I really | also very good at music making association's so I'm going | associations so i'm going to share with you an example that i really | Visa rhythms also very good and Messick making associations |
00:03:39,810 --> 00:03:44,780 [0.806000] | 00:03:39,220 --> 00:03:43,670 [1.396000] | 00:03:38,710 --> 00:03:40,616 [1.906000] | ||
example but I really like. A machine learning algorithm | to share with you an example that I really like sewing machine | so I'm going to share with you. | ||
00:03:40,790 --> 00:03:45,820 [3.411000] | 00:03:43,670 --> 00:03:49,640 [0.531000] | 00:03:42,743 --> 00:03:46,872 [1.458000] | 00:03:40,616 --> 00:03:44,201 [3.585000] | |
like %HESITATION so machine learning algorithm was asked to find | learning algorithm was asked to find associations between between | like so a machine learning algorithm was asked to find associations | An example that I really like, um, so uh machine learning | |
00:03:45,120 --> 00:03:49,140 [2.360000] | 00:03:45,820 --> 00:03:50,830 [1.660000] | 00:03:46,872 --> 00:03:52,338 [0.608000] | 00:03:44,201 --> 00:03:47,480 [3.279000] | |
was asked to find associations between | associations between %HESITATION between sorry between art works | between between sorry between artworks from britain from | algorithm was asked to find associations between. | |
00:03:49,140 --> 00:03:51,850 [4.185000] | 00:03:50,830 --> 00:03:56,050 [2.495000] | 00:03:49,640 --> 00:03:53,640 [3.685000] | 00:03:52,338 --> 00:03:57,369 [0.987000] | 00:03:48,340 --> 00:03:53,325 [4.985000] |
between - sorry - between art works from the... | from the dreaded from the Tate's collection of British art and Richard | sorry between artworks from the from the Tate's collection | the tate's collection of british art and images from two thousand | Um between sorry between art works from the bridge from |
00:03:51,850 --> 00:03:54,970 [3.120000] | 00:03:53,640 --> 00:04:00,130 [1.330000] | 00:03:53,325 --> 00:03:58,961 [1.645000] | ||
from the Tate's collection of British art | of British art and reuter images from 2016 and here is what | the Tates Collection of British art and rotor images from 2016 | ||
00:03:55,340 --> 00:03:58,860 [3.520000] | 00:03:56,160 --> 00:04:02,600 [2.700000] | 00:03:57,369 --> 00:04:06,487 [1.491000] | ||
And Reuters images from 2016. | images from two thousand sixteen and here is what the algorithm gave | sixteen and here is what algorithm gave us associations i think is | ||
00:03:58,860 --> 00:04:05,790 [6.930000] | 00:04:02,600 --> 00:04:09,470 [3.190000] | 00:04:00,130 --> 00:04:07,100 [5.660000] | 00:03:58,961 --> 00:04:01,670 [6.829000] | |
and here, is what the algorithm gave us. Associations. | us associations I think it's quite interesting with this work upright | the acronym gave us associations I think it's quite interesting | in here is what the algorithm. | |
00:04:02,400 --> 00:04:03,880 [1.480000] | ||||
Gave us associations. | ||||
00:04:05,790 --> 00:04:11,410 [5.620000] | 00:04:09,470 --> 00:04:18,880 [1.940000] | 00:04:08,290 --> 00:04:14,020 [3.120000] | 00:04:06,487 --> 00:04:13,046 [4.923000] | 00:04:05,820 --> 00:04:07,250 [5.590000] |
I think it's quite interesting. With this work, art work, | key %HESITATION type this a photograph from Reuters with this one she S. | with this work awkward qiyam type this photograph from borders | quite interesting with this what approach type this photograph | I think it's quite interesting. |
00:04:08,330 --> 00:04:10,590 [2.260000] | ||||
With this work out great Kia. | ||||
00:04:11,410 --> 00:04:15,990 [2.760000] | 00:04:13,046 --> 00:04:27,318 [1.124000] | 00:04:11,390 --> 00:04:14,170 [2.780000] | ||
he typed this photograph from Reuters. | from reuters he's associated with this win so that's kind of fun | Type this photograph from Reuters. | ||
00:04:15,990 --> 00:04:20,700 [4.710000] | 00:04:18,880 --> 00:04:28,280 [1.820000] | 00:04:15,990 --> 00:04:27,440 [4.710000] | 00:04:16,040 --> 00:04:17,040 [4.660000] | |
With this one, he associated | associated it's one so that's kind of fun but translated in | with this one chiasso Associated this one so that's kind of fun | With this one. | |
00:04:20,700 --> 00:04:27,580 [0.730000] | 00:04:18,380 --> 00:04:21,430 [3.050000] | |||
this one. So that's kind of fun but | He associated this one. | |||
00:04:27,580 --> 00:04:30,920 [1.984000] | 00:04:28,280 --> 00:04:32,610 [1.284000] | 00:04:27,440 --> 00:04:31,540 [2.124000] | 00:04:27,318 --> 00:04:32,152 [2.246000] | 00:04:26,230 --> 00:04:29,564 [3.334000] |
translated in the business world, these | the trend in the business world and these algorithms can use me | but translated in the trend in the business World these algorithms | but translated in the trend in the business world these algorithms | So that's kind of fun, but translated in the trend |
00:04:30,920 --> 00:04:35,770 [0.189000] | 00:04:29,564 --> 00:04:31,109 [1.545000] | |||
algorithms can use, the association ability of these | in the business world. | |||
00:04:32,610 --> 00:04:37,340 [2.646000] | 00:04:31,540 --> 00:04:37,340 [3.716000] | 00:04:32,152 --> 00:04:36,771 [3.104000] | 00:04:31,109 --> 00:04:35,256 [4.147000] | |
vis a vis the association ability of these algorithms can use to | can use me the association ability of these algorithms can use to | can use mean dvd association ability of these algorithms can | These algorithms can use I mean, the the Association ability | |
00:04:36,470 --> 00:04:38,510 [0.250000] | 00:04:35,256 --> 00:04:36,720 [1.464000] | |||
algorithms can use to segment - can be used to | of these algorithms. | |||
00:04:39,150 --> 00:04:45,600 [1.413000] | 00:04:37,340 --> 00:04:41,240 [3.223000] | 00:04:37,340 --> 00:04:41,240 [3.223000] | 00:04:36,771 --> 00:04:40,573 [3.792000] | 00:04:36,770 --> 00:04:40,563 [3.793000] |
segment consumers according to similar behaviors, for example. | segment can be used to segment consumers according to similar | segment can be used to segment consumers according to similar | use to segment can be used to segment consumers according to | Can use to segment can be used to segment consumers according |
00:04:41,240 --> 00:04:47,950 [1.480000] | 00:04:41,240 --> 00:04:48,360 [1.480000] | 00:04:40,573 --> 00:04:47,873 [2.147000] | 00:04:40,563 --> 00:04:42,720 [2.157000] | |
behaviors for example advances in the machine learning %HESITATION | behaviors for example adventures in the machine learning or not | similar behaviors for example advances in a machine learning | to similar behaviors for example? | |
00:04:45,600 --> 00:04:50,230 [4.630000] | 00:04:48,010 --> 00:04:51,750 [2.220000] | 00:04:48,360 --> 00:04:52,540 [1.870000] | 00:04:47,873 --> 00:04:52,298 [2.357000] | 00:04:45,660 --> 00:04:50,720 [4.570000] |
Advances in machine learning are not only due to the | are not only do it thanks to the advances in the algorithms | only thanks to the advancements in the algorithms themselves | are not only do it thanks to the advances in the algorithms | Advances in machine learning on not only do thanks to the advancements |
00:04:50,230 --> 00:04:53,910 [3.680000] | 00:04:51,750 --> 00:04:55,800 [2.160000] | 00:04:52,298 --> 00:04:57,607 [1.612000] | 00:04:50,720 --> 00:04:52,950 [3.190000] | |
advances in the algorithms themselves. | themselves %HESITATION these advances are due to two other | themselves these advances are due to two other factors that concurred | in the algorithms themselves. | |
00:04:53,910 --> 00:04:57,980 [4.463000] | 00:04:55,800 --> 00:05:00,090 [2.573000] | 00:04:53,910 --> 00:04:58,110 [4.463000] | 00:04:57,607 --> 00:05:01,778 [0.766000] | 00:04:53,880 --> 00:04:58,373 [4.493000] |
These advances are due to two other factors that concurred | factors that come Kurt %HESITATION at the same time that's about | these advances are due to two other factors that encourage at | at the same time that developed the same time when it's picked | These advances are due to 2 other factors that can curd at the same |
00:04:57,980 --> 00:05:01,180 [3.200000] | 00:05:00,090 --> 00:05:04,750 [1.090000] | 00:04:58,110 --> 00:05:02,050 [3.070000] | 00:04:58,373 --> 00:05:02,450 [2.807000] | |
at the same time. That developed at the same time. | the same time when it's big data %HESITATION with internet we've | the same time that developed the same time when it's picked | time that developed the same time, one, it's speak doctor. | |
00:05:01,180 --> 00:05:05,380 [4.200000] | 00:05:04,750 --> 00:05:08,940 [0.630000] | 00:05:03,230 --> 00:05:06,660 [2.150000] | 00:05:01,778 --> 00:05:06,690 [3.602000] | 00:05:03,950 --> 00:05:08,377 [1.430000] |
One is "big data". With internet we've been producing | been producing lots and lots of data in this data can be used | up with internet we've been producing lots and lots of data | up so with internet we've been producing lots and lots of data | With Internet we've been producing lots and lots of data and this data |
00:05:05,380 --> 00:05:09,580 [4.200000] | 00:05:09,180 --> 00:05:12,850 [0.400000] | 00:05:06,940 --> 00:05:11,690 [2.640000] | 00:05:06,970 --> 00:05:12,072 [2.610000] | 00:05:08,377 --> 00:05:10,820 [1.203000] |
lots and lots of data. And this data can be used to train | to train those algorithms to say the truth machine learning | and this data can be used to train those algorithms to say | and this that can be used to train those algorithms to say the truth | can be used to train those algorithms. |
00:05:09,580 --> 00:05:12,850 [3.270000] | 00:05:11,690 --> 00:05:14,340 [1.160000] | 00:05:12,072 --> 00:05:15,496 [0.778000] | 00:05:11,530 --> 00:05:14,640 [1.320000] | |
those algorithms. To say the truth, machine learning | the truth machine learning algorithms what are the idea of | machine learning algorithms were already idea of machine learning | Say, the truth machine learning algorithms or or | |
00:05:12,850 --> 00:05:15,960 [3.110000] | 00:05:12,850 --> 00:05:15,960 [3.110000] | 00:05:14,340 --> 00:05:18,380 [1.620000] | 00:05:15,496 --> 00:05:19,365 [0.464000] | 00:05:14,640 --> 00:05:16,537 [1.320000] |
algorithms or the idea of machine learning algorithms, | algorithms or the idea of machine learning algorithms %HESITATION | machine learning algorithms is he I mean has been developed for | algorithms is he i mean has been developed for a couple of years | the idea of machine learning. |
00:05:15,960 --> 00:05:19,580 [3.620000] | 00:05:15,960 --> 00:05:19,760 [3.620000] | 00:05:18,380 --> 00:05:21,350 [1.200000] | 00:05:19,365 --> 00:05:22,613 [0.215000] | 00:05:16,537 --> 00:05:19,951 [3.043000] |
has been developed for a couple of years or | is he I mean has been developed for a couple of years in in | a couple of years if you in the decades but we were liking | and in the decades but we were lucky that that to feed them | Algorithms is he has been, he developed for a couple of |
00:05:19,580 --> 00:05:22,530 [2.950000] | 00:05:19,760 --> 00:05:22,530 [2.770000] | 00:05:21,350 --> 00:05:25,890 [1.180000] | 00:05:19,951 --> 00:05:22,833 [2.579000] | |
even decades, but we were lacking the data to feed them | the decades but we would like to have the data to feed them | that that tattoo feed them enough so that was a big thing to do | years of even the decades, but we were liking. | |
00:05:22,940 --> 00:05:25,970 [4.410000] | 00:05:22,530 --> 00:05:27,350 [4.820000] | 00:05:25,890 --> 00:05:29,410 [1.460000] | 00:05:22,613 --> 00:05:27,497 [4.737000] | 00:05:22,833 --> 00:05:26,930 [4.517000] |
enough. So, that was a big thing to help | enough so that was a big thing to to help %HESITATION these | to help these ugly things take off and the second thing I needed | enough so that was a big thing to do to help these algorithms | That data to feed them enough so that was a big thing to to help him. |
00:05:26,280 --> 00:05:28,500 [2.220000] | 00:05:27,350 --> 00:05:30,880 [1.150000] | 00:05:27,497 --> 00:05:32,054 [1.003000] | 00:05:27,110 --> 00:05:30,443 [1.390000] | |
these algorithms takeoff and the | algorithms takeoff and the second thing I needed was computer | take off and the second thing he needed was computer power which | These algorithms take off and the second thing. | |
00:05:28,500 --> 00:05:32,310 [3.810000] | 00:05:31,110 --> 00:05:36,220 [1.200000] | 00:05:29,410 --> 00:05:35,670 [2.900000] | 00:05:32,054 --> 00:05:37,710 [0.256000] | 00:05:30,443 --> 00:05:34,777 [1.867000] |
second thing needed was, computer power, which now we | a power which now we have so basically today AI takes off | was computer a power which now we have so basically today AI | now we have so basically today takes off because the stars have aligned | He did was computer power which now we have so basically today. |
00:05:32,760 --> 00:05:34,270 [1.510000] | ||||
have. So basically, | ||||
00:05:34,850 --> 00:05:37,210 [3.010000] | 00:05:36,220 --> 00:05:37,680 [1.640000] | 00:05:35,670 --> 00:05:37,680 [2.190000] | 00:05:34,777 --> 00:05:37,860 [3.083000] | |
today AI takes off because the stars have | because the stars have aligned | takes off because the stars have aligned | AI takes off because the stars have aligned. | |
00:05:37,680 --> 00:05:38,890 [1.210000] | ||||
aligned. | ||||
00:05:38,890 --> 00:05:39,290 [0.400000] | ||||
And. |