0.25
0.5
0.75
1.25
1.5
1.75
2
Scientific Data Mining: Distilling Free-Form Natural Laws from Experimental Data
Published on Aug 29, 20114496 Views
For centuries, scientists have attempted to identify and document analytical laws that underlie physical phenomena in nature. Despite the prevalence of computing power, the process of finding natural
Related categories
Chapter list
Scientific Data Mining: Distilling Free-Form Natural Laws from Experimental Data00:00
Detect invariance00:52
Robotics (1)01:25
Robotics (2)02:11
Robotics (3)02:49
Robotics (4)03:42
Robotics (5)04:53
Robotics (6)05:07
Robotics (7)05:46
Robotics (8)06:29
Adapting in simulation07:11
Adapting in reality07:45
Simulation & Reality08:10
Servo actuators / Tilt sensors09:46
Self-Model / Explanatory Action / Target Behaviour synthesis11:24
Morphological Estimation12:42
Emergent Self-Model13:16
Damage Recovery14:39
Random / Predicted / Physical15:30
Theory of mind16:08
System Identification16:23
Inference Process17:08
Bridge (1)17:58
Bridge (2)18:10
Static ID: Damage Diagnosis18:24
Discrete Dynamics Inference18:56
Circuit Building Blocks19:13
Symbolic Regression19:55
Encoding Equations21:50
Models / Experiments22:17
Models22:58
Solution Accuracy23:09
Solution Complexity23:28
Stock market board24:14
Prime Numbers24:30
Semi-empirical mass formula25:00
Boolean Ellipse26:15
Systems of Differential Equations26:50
Inferring Biological Networks26:55
Wet Data, Unknown System28:12
Data (1)28:41
Data (2)29:12
Symbolic Regression30:40
Withheld Test Set #1 Fit31:52
Withheld Test Set #2 Fit31:58
Withheld Test Set #3 Fit31:59
Looking For Invariants32:06
Data Mining32:39
Pendulum33:22
4235:00
From Data / From Equation36:17
Experiments (1)36:53
Experiments (2)37:11
Experiments (3)37:27
Experiments (4)37:58
Double Linear Oscillator (1)38:13
Double Linear Oscillator (2)38:33
Double Linear Oscillator (3)38:45
Double Linear Oscillator (4)38:48
Double Linear Oscillator (5)39:13
A have a lot of data ...39:27
Software online: Eureqa39:50
Eureqa40:15
Citizen science40:35
Radon detection40:56
Strength of welded joints41:22
Monkey41:40
Scalability41:59
Approximations42:54
Alphabet43:53
Time to Regress44:16
Stohastic Models44:29
10% Noise44:41
30% Noise44:54
70% Noise44:58
Train / Validation Data45:18
Regressing Stochastic elements45:34
Simulation46:22
Likelihood Fitness48:24
Sample the Timespan (1)48:34
Sample the Timespan (2)48:41
Sample the Timespan (3)49:00
Concluding Remarks (1)49:14
Concluding Remarks (2)49:55