0.25
0.5
0.75
1.25
1.5
1.75
2
Reasoning, Attention and Memory
Published on Aug 23, 201613393 Views
The machine learning community has had great success in the last decades at solving basic prediction tasks such as text classification, image annotation and speech recognition. However, solutions to d
Related categories
Chapter list
Reasoning, Attention and Memory00:00
Deep Learning for Vision01:20
Deep Learning for Speech01:59
Deep Learning for Text02:52
Deep Models - 103:31
Deep Models - 203:59
Scenario 104:06
Scenario 205:25
Scenario 3 - 106:56
Scenario 3 - 207:23
Scenario 407:32
What is Required? - 108:22
What is Required? - 209:48
Possible Solution10:02
Outline11:19
General Architecture12:38
Memory Networks - 114:48
Memory Networks - 216:14
Memory Networks (Fully Supervised) - 117:30
Memory Networks (Fully Supervised) - 218:47
Memory Networks (Fully Supervised) - 319:54
Memory Networks (Fully Supervised) - 420:07
Memory Networks (Fully Supervised) - 521:39
Memory Networks (Fully Supervised) - 622:07
Memory Networks (Fully Supervised) - 722:50
Memory Networks (Fully Supervised) - 823:20
Memory Networks (Fully Supervised) - 924:51
Memory Networks (Fully Supervised) - 1025:27
Memory Networks (Fully Supervised) - 1126:12
bAbI Dataset: Slight Digression26:28
bAbI Dataset: Simulator - 127:46
bAbI Dataset: Simulator - 228:10
bAbI Dataset - 129:14
bAbI Dataset - 229:37
bAbI Dataset - 329:50
bAbI Dataset - 430:32
bAbI Dataset - 530:51
bAbI Dataset - 630:55
bAbI Dataset - 731:23
bAbI Dataset - 831:46
bAbI Dataset - 932:13
bAbI Dataset - 1032:26
bAbI Dataset - 1132:30
bAbI Dataset - 1232:33
MemNNs on bAbI - 132:35
MemNNs on bAbI - 233:49
MemNNs on bAbI - 334:48
MemNNs on bAbI - 435:28
MemNNs on bAbI - 535:46
MemNNs on bAbI - 635:53
Full Supervision in MemNNs - 136:14
Full Supervision in MemNNs - 241:28
End2End MemNNs - 141:40
End2End MemNNs - 348:44
End2End MemNNs - 250:17
E2EMemNNs: Other Details51:38
E2EMemNNs: bAbI - 151:56
E2EMemNNs: bAbI - 252:28
E2EMemNNs: Language Modeling - 153:23
E2EMemNNs: Language Modeling - 255:35
Relevant Literature56:08
Large Scale Memories - 157:28
Large Scale Memories - 257:42
Reverb Dataset57:58
MemNNs on Reverb Dataset - 158:28
MemNNs on Reverb Dataset - 258:43
Multitasked MemNNs:bAbI + Reverb59:15
Cloze Style QA01:00:09
CBT: Children’s Book Dataset - 101:01:42
CBT: Children’s Book Dataset - 201:02:16
MemNNs for Story Understanding - 101:03:08
MemNNs for Story Understanding - 201:03:30
MemNNs for Story Understanding - 301:03:47
MemNNs for Story Understanding - 401:04:46
Self Supervision in MemNNs01:04:54
QA on News Articles - 101:04:56
QA on News Articles - 201:05:35
QA on News Articles - 301:06:07
Dialog Modeling - 101:06:17
Dialog Modeling - 201:06:24
Dialog Modeling - 301:06:32
Dialog Modeling - 401:06:34
Dialog Modeling - 501:06:38
Memory Networks for Dialog01:06:41
Results01:07:16
Key-Value MemNNs - 101:07:32
Key-Value MemNNs - 201:07:54
Key-Value MemNNs - 301:08:20
Dynamic MemNNs - 101:08:22
Dynamic MemNNs - 201:08:49
Dynamic MemNNs - 301:09:59
Dynamic MemNNs - 401:10:00
Dynamic MemNNs Experiments - 101:10:02
Dynamic MemNNs Experiments - 201:10:11
Dynamic MemNNs Experiments - 301:10:18
MemNNs Summary01:10:23
MemNNs Shortcomings01:10:55
Neural Turing Machines01:11:43
NTM: Read Mechanism01:12:03
NTM: Write Mechanism01:12:37
NTM: Addressing Mechanism - 101:13:56
NTM: Addressing Mechanism - 201:14:23
NTM: Addressing Mechanism - 301:14:50
NTM: Experiments - 101:16:28
NTM: Experiments - 201:17:19
NTM: Experiments - 301:17:20
NTM: Experiments - 401:17:47
NTM: Experiments - 501:17:48
NTM: Experiments - 601:18:08
NTM: Summary01:18:18
Stack Augmented RNNs - 101:19:03
Stack Augmented RNNs - 201:19:10
Stack Augmented RNNs - 301:19:10
Stack Augmented RNNs - 401:19:11
Stack Augmented RNNs - 501:19:12
Wrapping Up - 101:19:12
Wrapping Up - 201:19:41
Thank You01:20:40