Timeline for Relationship between information retrieval and source separation in signal processing
Current License: CC BY-SA 3.0
27 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 12, 2016 at 17:08 | vote | accept | Srishti M | ||
| Jul 10, 2016 at 19:01 | history | edited | Srishti M | CC BY-SA 3.0 |
updated Question, provided more details; title elaborated
|
| Jul 8, 2016 at 6:47 | history | edited | Srishti M | CC BY-SA 3.0 |
updated Question, provided more details
|
| Jul 8, 2016 at 6:35 | comment | added | Srishti M | In my question I say $\mathbf{y}$ is of $ L<<D$ because this vector is unknown and needs to be estimated such that the target $\mathbf{y} $ is equal to the input $\mathbf{x}$. I will put these information as well. | |
| Jul 8, 2016 at 6:35 | comment | added | Srishti M | It seems that the training and the testing examples are subjected to some transformation that involves dimensional reduction where the features/ input is represented in a reduced feature space. The recognition and perhaps the learning is done in this transformed feature space. Due to this block, it is not correct to say that $\mathbf{z = y}$ because $z_n$ is the $L$ number of bits and $\mathbf{y}$ is the recognized image - the desired target corresponding to a query as an input in the recognition / retrieval task. | |
| Jul 8, 2016 at 6:29 | comment | added | Srishti M | Under the Edit, the image contains three blocks : (a),(b), and (c). The input $\mathbf{x}$ is of D bits and that is is fed into (b) which acts as the mapping function - the encoder $h(x) : \mathbf{x} \mapsto \mathbf{z}$. It appears that the input is transformed into $L<<D$ bits. This I called as $\mathbf{y}$ that is the intermediate block similar to dimensionality reduction from $D$ bits to $L$ bits. The mapping function $f()$ regenerates the target output from $L$ bits. | |
| Jul 8, 2016 at 4:41 | comment | added | robert bristow-johnson | okay, you still have at the top $\mathbf{y}$ a vector with $L$ elements. but at the bottom you say that $\mathbf{y}$ a vector with $D$ elements. and in Eq. (2) it also appears that $\mathbf{x}$ and $\mathbf{y}$ have the same number of elements. and where does the variable $z_n$ come from? why is $z_n$ not a $y_n$? | |
| Jul 8, 2016 at 4:32 | history | edited | robert bristow-johnson | CC BY-SA 3.0 |
one very small fix
|
| Jul 8, 2016 at 3:47 | history | edited | Srishti M | CC BY-SA 3.0 |
added 142 characters in body; edited tags
|
| Jul 8, 2016 at 3:35 | history | edited | Srishti M | CC BY-SA 3.0 |
corrected the notations; edited the Question
|
| Jul 8, 2016 at 2:38 | comment | added | Gilles | I agree with @robertbristow-johnson that you should clarify most of the notations. Especially the $y$, $\mathbf y$, $x$, $\mathbf x$, $h$, and $\mathbf h$. And if possible please add some text to the diagram in your EDIT, right now it's just there. | |
| Jul 8, 2016 at 2:13 | comment | added | robert bristow-johnson | too late to modify. in my comment above, i think replace "$L$" with "$D$". or is $\mathbf{w}$ a $L \times D$ matrix? | |
| Jul 8, 2016 at 2:10 | comment | added | robert bristow-johnson | And how, exactly, is "$y$" related to "$\mathbf{y}$"? Is, somehow, $y$ the elements of vector $\mathbf{y}$? | |
| Jul 8, 2016 at 2:05 | history | edited | Gilles | CC BY-SA 3.0 |
improved formatting for paragraphs, list of questions, etc
|
| Jul 8, 2016 at 2:03 | comment | added | robert bristow-johnson | if "the channel" is a "Moving Average", then $\mathbf{w} = \{1/L, 1/L, ... 1/L\}$ and $h(\cdot)$ defined similarly, i think. but that seems to conflict with some other optimal definition of $\mathbf{w}$ or $h(\cdot)$ for whatever purpose of classification into $y_1$ or $y_2$. so it's still not clear to me exactly how the problem is defined. | |
| Jul 8, 2016 at 1:49 | comment | added | robert bristow-johnson | What, exactly is "$x$"? (not "$\mathbf{x}$" nor "$x_n$".) then the next question is: What is "$h(x)$"?, but that might get answered consequentially from answering the first question. | |
| Jul 8, 2016 at 1:44 | history | edited | robert bristow-johnson | CC BY-SA 3.0 |
slight formatting adjustment.
|
| Jul 8, 2016 at 1:37 | comment | added | robert bristow-johnson | i friggin' up-arrowed this just based on the title. now i'm gonna read it. | |
| Jul 8, 2016 at 1:16 | history | edited | Srishti M | CC BY-SA 3.0 |
added 599 characters in body
|
| Jul 8, 2016 at 1:05 | history | edited | Srishti M | CC BY-SA 3.0 |
improved Question
|
| Jul 7, 2016 at 23:56 | history | edited | Gilles | CC BY-SA 3.0 |
formatting listed points in EDIT
|
| Jul 7, 2016 at 23:32 | history | edited | Srishti M | CC BY-SA 3.0 |
updated Question, provided more details
|
| Jul 7, 2016 at 11:22 | history | tweeted | twitter.com/StackSignals/status/751013404354183168 | ||
| Jul 7, 2016 at 8:58 | answer | added | A_A | timeline score: 5 | |
| Jul 7, 2016 at 3:15 | history | edited | Srishti M | CC BY-SA 3.0 |
fixed typo
|
| Jul 7, 2016 at 2:54 | history | edited | Gilles | CC BY-SA 3.0 |
improved formatting
|
| Jul 6, 2016 at 21:41 | history | asked | Srishti M | CC BY-SA 3.0 |