[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

• ÚFAL Wiki Home
• ÚFAL Home
• Sitemap
• RSS
• ÚFAL calendar
• Recent changes
• Login
• Hledat:

[ Back to the navigation ]

Differences

This shows you the differences between two versions of the page.

View differences: Side by SideInline Go

Link to this comparison view

Both sides previous revision Previous revision 2014/10/12 15:04 popel 2014/10/12 15:03 popel 2013/03/27 16:56 vandemos 2013/03/22 14:54 popel 2013/03/19 00:05 popel 2013/03/18 20:27 vandemos vytvořeno Go Next revision		Previous revision 2014/10/12 15:04 popel 2014/10/12 15:03 popel 2013/03/27 16:56 vandemos 2013/03/22 14:54 popel 2013/03/19 00:05 popel 2013/03/18 20:27 vandemos vytvořeno Go
courses:rg:2013:memm [2014/10/12 15:03] popel		courses:rg:2013:memm [2014/10/12 15:04] (current) popel
Line 1:		Line 1:
	===== Maximum Entropy Markov Models - Questions =====		===== Maximum Entropy Markov Models - Questions =====
-	**1. Explain (roughly) how the new formula for α_t+1(s) is derived (i.e. formula 1 in the paper).**	+	1. Explain (roughly) how the new formula for α_t+1(s) is derived (i.e. formula 1 in the paper).
-	**2. Section 2.1 states "we will split P(s|s',o) into |S| separately trained transition functions". What are the advantages and disadvantages of this approach?**	+	2. Section 2.1 states "we will split P(s|s',o) into |S| separately trained transition functions". What are the advantages and disadvantages of this approach?
-	**3. Let S= {V,N} (verb and non-verb)	+	3. Let S= {V,N} (verb and non-verb)
	Training data = he/N can/V can/V a/N can/N		Training data = he/N can/V can/V a/N can/N
	//Observation features// are:		//Observation features// are:
Line 12:		Line 12:
	b3 = current word is “a” and next word is “can”		b3 = current word is “a” and next word is “can”
	When implementing MEMM you need to define s_0, i.e. the previous state before the first token. It may be a special NULL, but for simplicity let’s define it as N.		When implementing MEMM you need to define s_0, i.e. the previous state before the first token. It may be a special NULL, but for simplicity let’s define it as N.
-	a) What are the states (s) and observations (o) for this training data?**	+	a) What are the states (s) and observations (o) for this training data?
-	**b) Equation (2) defines features f_a based on //observation features// b. How many such f_a features do we have?**	+	b) Equation (2) defines features f_a based on //observation features// b. How many such f_a features do we have?
-	**c) Equation (3) defines constraints. How many such constraints do we have?**	+	c) Equation (3) defines constraints. How many such constraints do we have?
-	**d) List all the constraints involving feature b2, i.e. substitute (whenever possible) concrete numbers into Equation (3).**	+	d) List all the constraints involving feature b2, i.e. substitute (whenever possible) concrete numbers into Equation (3).
-	**e) In step 3 of the GIS algorithm you need to compute <latex>P_{s’}^{(j)}(s|o)</latex>. Compute <latex>P_N^{(0)}(N|can)</latex> and <latex>P_N^{(0)}(V|can)</latex>.**	+	e) In step 3 of the GIS algorithm you need to compute <latex>P_{s’}^{(j)}(s|o)</latex>. Compute <latex>P_N^{(0)}(N|can)</latex> and <latex>P_N^{(0)}(V|can)</latex>.
	**Hint** : You might be confused about the m_s' variable (and  t_1, …, <latex>t_{m_{s'}}</latex>) in Equation (3).		**Hint** : You might be confused about the m_s' variable (and  t_1, …, <latex>t_{m_{s'}}</latex>) in Equation (3).

---

[ Back to the navigation ] [ Back to the content ]