▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
tr ts ts t - - PowerPoint PPT Presentation
tr ts ts t - - PowerPoint PPT Presentation
tr ts t t tr ts ts t
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❙✉♠♠❛r②
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
Pr♦❜❧❡♠
❨♦✉ ❛r❡ ❢❛❝❡❞ r❡♣❡❛t❡❞❧② ✇✐t❤ ❛ ❝❤♦✐❝❡ ❛♠♦♥❣ n ❞✐✛❡r❡♥t ♦♣t✐♦♥s✳ ❆❢t❡r ❡❛❝❤ ❝❤♦✐❝❡ ②♦✉ r❡❝❡✐✈❡ ❛ ♥✉♠❡r✐❝❛❧ r❡✇❛r❞✳ ❨♦✉r ♦❜❥❡❝t✐✈❡ ✐s t♦ ♠❛①✐♠✐③❡ t❤❡ ❡①♣❡❝t❡❞ t♦t❛❧ r❡✇❛r❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❙❧♦t ♠❛❝❤✐♥❡ ❛r❡ ❦♥♦✇♥ ❛s
♦♥❡✲❛r♠❡❞ ❜❛♥❞✐ts✱ ❜❡❝❛✉s❡ t❤❡② ✇❡r❡ ♦r✐❣✐♥❛❧❧② ♦♣❡r❛t❡❞ ❜② ♦♥❡ ❧❡✈❡r ♦♥ t❤❡ s✐❞❡ ♦❢ t❤❡ ♠❛❝❤✐♥❡✳
- ❆ ❣❛♠❜❧❡r str❛t❡❣✐❝❛❧❧②
♦♣❡r❛t✐♥❣ ♠✉❧t✐♣❧❡ ♠❛❝❤✐♥❡s ✐♥ ♦r❞❡r t♦ ❞r❛✇ t❤❡ ❤✐❣❤❡st ♣♦ss✐❜❧❡ ♣r♦✜ts ✐s ❝❛❧❧❡❞ ❛ ♠✉❧t✐✲❛r♠❡❞ ❜❛♥❞✐t✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❚❤❡ r❡✇❛r❞ ❢♦r ❡❛❝❤ ♦♥❡✲❛r♠❡❞ ❜❛♥❞✐t ❤❛s ❛ ❞✐✛❡r❡♥t
❞✐str✐❜✉t✐♦♥✱ t❤❡r❡❢♦r❡ ❛ ❞✐✛❡r❡♥t ❡①♣❡❝t❛t✐♦♥✳
- ■❢ t❤❡ ❡①♣❡❝t❛t✐♦♥s ✇❡r❡ ❦♥♦✇♥✱ ✐t ✇♦✉❧❞ ❜❡ tr✐✈✐❛❧ t♦ s♦❧✈❡ t❤❡
n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✿ s❡❧❡❝t t❤❡ ❛❝t✐♦♥ ✇✐t❤ t❤❡ ❤✐❣❤❡st ♠❡❛♥ r❡✇❛r❞✳
- ❚❤✐s ✐s ✇❤② ✇❡ ❛ss✉♠❡ t❤❛t ✇❡ ❞♦♥✬t ❦♥♦✇ t❤❡ ❡①♣❡❝t❡❞
r❡✇❛r❞s✱ ❛❧t❤♦✉❣❤ ✇❡ ♠✐❣❤t ❤❛✈❡ ❡st✐♠❛t❡s✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❊①♣❧♦r✐♥❣ ❛♥❞ ❡①♣❧♦✐t✐♥❣ ♣r♦❜❧❡♠
- ❆t ❛♥② t✐♠❡ st❡♣ t❤❡r❡ ✐s ❛t ❧❡❛st ♦♥❡ ❛❝t✐♦♥ ✇❤♦s❡ ❡st✐♠❛t❡❞
❡①♣❡❝t❡❞ r❡✇❛r❞ ✐s ❣r❡❛t❡st✳ ❲❡ ❝❛❧❧ t❤✐s ❛ ❣r❡❡❞② ❛❝t✐♦♥✳
- ■❢ ②♦✉ s❡❧❡❝t ❛ ❣r❡❡❞② ❛❝t✐♦♥✱ ✇❡ s❛② t❤❛t ②♦✉ ❛r❡ ❡①♣❧♦✐t✐♥❣
②♦✉r ❝✉rr❡♥t ❦♥♦✇❧❡❞❣❡ ♦❢ t❤❡ ✈❛❧✉❡s ♦❢ t❤❡ ❛❝t✐♦♥s✳
- ■❢ ✐♥st❡❛❞ ②♦✉ s❡❧❡❝t ♦♥❡ ♦❢ t❤❡ ♥♦♥✲❣r❡❡❞② ❛❝t✐♦♥s✱ t❤❡♥ ✇❡
s❛② ②♦✉ ❛r❡ ❡①♣❧♦r✐♥❣✱ ❜❡❝❛✉s❡ t❤✐s ❡♥❛❜❧❡s ②♦✉ t♦ ✐♠♣r♦✈❡ ②♦✉r ❡st✐♠❛t❡ ♦❢ t❤❡ ♥♦♥✲❣r❡❡❞② ❛❝t✐♦♥✬s ❡①♣❡❝t❡❞ r❡✇❛r❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❖♥❡✲❛r♠❡❞ ❜❛♥❞✐t
❲❡ ❞❡♥♦t❡ t❤❡ tr✉❡ ♠❡❛♥ r❡✇❛r❞ ♦❢ ❛♥ ❛❝t✐♦♥ a ❛s q(a)✱ ❛♥❞ t❤❡ ❡st✐♠❛t❡❞ ♠❡❛♥ r❡✇❛r❞ ♦♥ t❤❡ t✲t❤ t✐♠❡ st❡♣ ❛s Qt(a)✳
❊st✐♠❛t♦r ❢♦r t❤❡ ♠❡❛♥ r❡✇❛r❞ ✭s❛♠♣❧❡✲❛✈❡r❛❣❡✮
■❢ ❜② t❤❡ t✲t❤ t✐♠❡ st❡♣ ❛❝t✐♦♥ a ❤❛s ❜❡❡♥ ❝❤♦s❡♥ Nt(a) t✐♠❡s ♣r✐♦r t♦ t✱ ②✐❡❧❞✐♥❣ r❡✇❛r❞s R✶, R✷, . . . , RNt(a)✱ t❤❡♥ ✐ts ✈❛❧✉❡ ✐s ❡st✐♠❛t❡❞ t♦ ❜❡ Qt(a) = R✶ + R✷ + · · · + RNt(a) Nt(a) ✭✶✮
▲❛✇ ♦❢ ❧❛r❣❡ ♥✉♠❜❡rs
❆s Nt(a) → ∞✱ Qt(a) ❝♦♥✈❡r❣❡s t♦ q(a)✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆❝t✐♦♥✲s❡❧❡❝t✐♦♥ r✉❧❡s
- ❚❤❡ ❣r❡❡❞② ❛❝t✐♦♥ s❡❧❡❝t✐♦♥ ♠❡t❤♦❞✿
At = ❛r❣♠❛①
a
Qt(a) ✭✷✮
- ❚❤❡ ǫ✲❣r❡❡❞② ♠❡t❤♦❞s✿ ❜❡❤❛✈❡ ❣r❡❡❞✐❧② ♠♦st ♦❢ t❤❡ t✐♠❡✱ ❜✉t
❡✈❡r② ♦♥❝❡ ✐♥ ❛ ✇❤✐❧❡✱ s❛② ✇✐t❤ s♠❛❧❧ ♣r♦❜❛❜✐❧✐t② ǫ✱ s❡❧❡❝t r❛♥❞♦♠❧② ❢r♦♠ ❛♠♦♥❣st ❛❧❧ t❤❡ ❛❝t✐♦♥s ✇✐t❤ ❡q✉❛❧ ♣r♦❜❛❜✐❧✐t② ✐♥❞❡♣❡♥❞❡♥t❧② ♦❢ t❤❡ ❛❝t✐♦♥ ♠❡❛♥ r❡✇❛r❞s ❡st✐♠❛t❡s ✭❛❞✈❛♥t❛❣❡✿ ❛s Nt(a) → ∞✱ ✇❡ ❡♥s✉r❡ t❤❛t Qt(a) ❝♦♥✈❡r❣❡ t♦ q(a)✮✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❋✐❣✉r❡✿ ❆✈❡r❛❣❡ ♣❡r❢♦r♠❛♥❝❡ ♦❢ ǫ✲❣r❡❡❞② ❛❝t✐♦♥✲✈❛❧✉❡ ♠❡t❤♦❞s ♦♥ t❤❡ ✶✵✲❛r♠❡❞ t❡st❜❡❞✳ ❚❤❡s❡ ❞❛t❛ ❛r❡ ❛✈❡r❛❣❡s ♦✈❡r ✷✵✵✵ t❛s❦s✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
■♥❝r❡♠❡♥t❛❧ ■♠♣❧❡♠❡♥t❛t✐♦♥
- ❙♦ ❢❛r ✇❡ ❡st✐♠❛t❡❞ t❤❡ ❛❝t✐♦♥ ♠❡❛♥ r❡✇❛r❞s ❛s s❛♠♣❧❡
❛✈❡r❛❣❡s ♦❢ ♦❜s❡r✈❡❞ r❡✇❛r❞s✳
- Pr♦❜❧❡♠✿ t❤❡ ♠❡♠♦r② ❛♥❞ ❝♦♠♣✉t❛t✐♦♥❛❧ r❡q✉✐r❡♠❡♥ts ❢♦r t❤❡
✐♠♣❧❡♠❡♥t❛t✐♦♥ ❣r♦✇ ♦✈❡r t✐♠❡ ✇✐t❤♦✉t ❜♦✉♥❞✳
- ❙♦❧✉t✐♦♥✿ ❧❡t Qk ❞❡♥♦t❡ t❤❡ ❡st✐♠❛t❡ ❢♦r t❤❡ k✲t❤ r❡✇❛r❞✱ t❤❛t
✐s t❤❡ ❛✈❡r❛❣❡ ♦❢ ✐ts ✜rst k − ✶ r❡✇❛r❞s✱ ❛♥❞ ❛ k✲t❤ r❡✇❛r❞ ❢♦r t❤❡ ❛❝t✐♦♥✱ Rk✳ ❚❤❡♥✿ Qk+✶ = Qk + ✶ k [Rk − Qk] ✭✸✮
- ●❡♥❡r❛❧ ❢♦r♠✿
◆❡✇ ❊st✐♠❛t❡ ← ❖❧❞❊st✐♠❛t❡ + ❙t❡♣❙✐③❡[❚❛r❣❡t − ❖❧❞❊st✐♠❛t❡]
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
◆♦♥✲st❛t✐♦♥❛r② Pr♦❜❧❡♠
- ❙♦ ❢❛r✿ st❛t✐♦♥❛r② ❡♥✈✐r♦♥♠❡♥t❀ t❤❡ ❜❛♥❞✐t ✐s ♥♦t ❝❤❛♥❣✐♥❣
♦✈❡r t✐♠❡✳
- ■♥ ♣r❛❝t✐❝❡✱ t❤✐s ❤②♣♦t❤❡s✐s ✐s ✈❡r② ♦❢t❡♥ ✈✐♦❧❛t❡❞ ♦r ✐♠♣♦ss✐❜❧❡
t♦ ✈❡r✐❢②✱ ❛♥❞ ✇❡ ❡♥❝♦✉♥t❡r ♥♦♥✲st❛t✐♦♥❛r✐t②✳
- ■♥ s✉❝❤ ❝❛s❡s ✇❡✐❣❤t r❡❝❡♥t r❡✇❛r❞s ♠♦r❡ ❤❡❛✈✐❧② t❤❛♥
❧♦♥❣✲♣❛st ♦♥❡s✳
- Qk+✶ = Qk + α[Rk − Qk]
✭✹✮ ✇❤❡r❡ t❤❡ st❡♣✲s✐③❡ ♣❛r❛♠❡t❡r α ∈ (✵, ✶] ✐s ❝♦♥st❛♥t✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
◆♦♥✲st❛t✐♦♥❛r② Pr♦❜❧❡♠ ✲ ✷
- ❲❡✐❣❤t❡❞ ❛✈❡r❛❣❡ ♦❢ t❤❡ ♣❛st r❡✇❛r❞s ❛♥❞ ♦❢ t❤❡ ✐♥✐t✐❛❧
❡st✐♠❛t❡ Q✶✿ Qk+✶ = (✶ − α)kQ✶ +
k
- i=✶
α(✶ − α)k−iRi ✭✺✮
- ❚❤❡ q✉❛♥t✐t② ✶ − α ✐s ❧❡ss t❤❛♥ ✶✱ ❛♥❞ t❤✉s t❤❡ ✇❡✐❣❤t ❣✐✈❡♥ t♦
Ri ❞❡❝r❡❛s❡s ❛s t❤❡ ♥✉♠❜❡r ♦❢ ✐♥t❡r✈❡♥✐♥❣ r❡✇❛r❞s ✐♥❝r❡❛s❡s✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❯♣♣❡r✲❈♦♥✜❞❡♥❝❡✲❇♦✉♥❞ ❆❝t✐♦♥ ❙❡❧❡❝t✐♦♥
- ■♥ t❤❡ ǫ✲❣r❡❡❞② ♠❡t❤♦❞✱ ✇❡ ❝❤♦♦s❡ ❛♥♦t❤❡r ❛❝t✐♦♥ ✭t❤❛t ✐s ♥♦t
t❤❡ ❣r❡❡❞② ♦♥❡✮✱ ✇✐t❤ ♣r♦❜❛❜✐❧✐t② ǫ✳
- ❇❡tt❡r ✇♦✉❧❞ ❜❡ t♦ s❡❧❡❝t ❛♠♦♥❣ t❤❡ ♥♦♥✲❣r❡❡❞② ❛❝t✐♦♥s t❤❡
♦♥❡ ✇❤✐❝❤ ❤❛s t❤❡ ❤✐❣❤❡st ♣r♦❜❛❜✐❧✐t② t♦ ❜❡ t❤❡ ❣r❡❡❞② ❛❝t✐♦♥ ✭✇✐t❤ t❤❡ ❜✐❣❣❡st ❡①♣❡❝t❛t✐♦♥ ♦❢ r❡✇❛r❞✮✿ At = ❛r❣♠❛①
a
- Qt(a) + c
- ❧♥t
Nt(a)
- ✭✻✮
✇❤❡r❡ c > ✵ ❝♦♥tr♦❧s t❤❡ ❞❡❣r❡❡ ♦❢ ❡①♣❧♦r❛t✐♦♥✳ ■❢ Nt(a) = ✵✱ t❤❡♥ a ✐s ❝♦♥s✐❞❡r❡❞ t♦ ❜❡ ❛ ♠❛①✐♠✐s✐♥❣ ❛❝t✐♦♥✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❙❡t ˆ
µ✶ = · · · = ˆ µk = ✵✱ n✶ = · · · = nk = ✵
- ❋♦r t = ✶ : T✿
- ❋♦r ❡❛❝❤ ❛r♠ i ❝❛❧❝✉❧❛t❡ UCB(i) = ˆ
µi + ✷
- ❧♥ t
ni
- P✐❝❦ ❛r♠ j = ❛r❣♠❛①
i
UCB(i) ❛♥❞ ♦❜s❡r✈❡ yt
- ❙❡t nj ← nj + ✶ ❛♥❞ ˆ
µj ← ˆ µj + ✶
nj (yt − ˆ
µj)✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❋✐❣✉r❡✿ ❆✈❡r❛❣❡ ♣❡r❢♦r♠❛♥❝❡ ♦❢ ❯❈❇ ❛❝t✐♦♥ s❡❧❡❝t✐♦♥ ♦♥ t❤❡ ✶✵✲❛r♠❡❞ t❡st❜❡❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- r❛❞✐❡♥t ❇❛♥❞✐ts
- ❙♦ ❢❛r✱ ✇❡ ❝♦♥s✐❞❡r❡❞ ♠❡t❤♦❞s t❤❛t ❡st✐♠❛t❡ t❤❡ ♠❡❛♥ r❡✇❛r❞
♦❢ ❛♥ ❛❝t✐♦♥ ❛♥❞ ✉s❡ t❤♦s❡ ❡st✐♠❛t♦rs t♦ s❡❧❡❝t ❛♥ ❛❝t✐♦♥✳
- ❍❡r❡ ✇❡ ❝♦♥s✐❞❡r ❧❡❛r♥✐♥❣ ❛ ♥✉♠❡r✐❝❛❧ ♣r❡❢❡r❡♥❝❡ Ht(a) ❢♦r
❡❛❝❤ ❛❝t✐♦♥ a✳ ❚❤❡ ❧❛r❣❡r t❤❡ ♣r❡❢❡r❡♥❝❡✱ t❤❡ ♠♦r❡ ♦❢t❡♥ t❤❛t ❛❝t✐♦♥ ✐s t❛❦❡♥✱ ❜✉t t❤❡ ♣r❡❢❡r❡♥❝❡ ❤❛s ♥♦ ✐♥t❡r♣r❡t❛t✐♦♥ ✐♥ t❡r♠s ♦❢ r❡✇❛r❞✿ P[At = a] = eHt(a) n
b=✶ eHt(b) = πt(a).
✭✼✮
- ■♥✐t✐❛❧❧② ❛❧❧ ♣r❡❢❡r❡♥❝❡s ❛r❡ t❤❡ s❛♠❡ ✭❡✳❣✳ H✶(a) = ✵✱ ∀a✮✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- r❛❞✐❡♥t ❇❛♥❞✐ts ✲ ✷
- ❖♥ ❡❛❝❤ st❡♣✱ ❛❢t❡r s❡❧❡❝t✐♥❣ t❤❡ ❛❝t✐♦♥ At ❛♥❞ r❡❝❡✐✈✐♥❣
r❡✇❛r❞ Rt✱ t❤❡ ♣r❡❢❡r❡♥❝❡s ❛r❡ ✉♣❞❛t❡❞ ❜②✿ Ht+✶(At) = Ht(At) + α(Rt − ¯ Rt)(✶ − πt(At)), ❛♥❞ Ht+✶(a) = Ht(a) − α(Rt − ¯ Rt)πt(a), ∀a = At ✭✽✮ ✇❤❡r❡ α > ✵ ✐s ❛ st❡♣✲s✐③❡ ♣❛r❛♠❡t❡r✱ ❛♥❞ ¯ Rt ∈ R ✐s t❤❡ ❛✈❡r❛❣❡ ♦❢ ❛❧❧ t❤❡ r❡✇❛r❞s ✉♣ t❤r♦✉❣❤ ❛♥❞ ✐♥❝❧✉❞✐♥❣ t✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❋✐❣✉r❡✿ ❆✈❡r❛❣❡ ♣❡r❢♦r♠❛♥❝❡ ♦❢ t❤❡ ❣r❛❞✐❡♥t✲❜❛♥❞✐t ❛❧❣♦r✐t❤♠ ♦♥ t❤❡ ✶✵✲❛r♠❡❞ t❡st❜❡❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❚❤❡r❡ ❛r❡ t❤r❡❡ ❢✉♥❞❛♠❡♥t❛❧ ❢♦r♠❛❧✐③❛t✐♦♥s ♦❢ t❤❡ ❜❛♥❞✐t
♣r♦❜❧❡♠ ❞❡♣❡♥❞✐♥❣ ♦♥ t❤❡ ❛ss✉♠❡❞ ♥❛t✉r❡ ♦❢ t❤❡ r❡✇❛r❞ ♣r♦❝❡ss✿ st♦❝❤❛st✐❝✱ ❛❞✈❡rs✐❛❧ ❛♥❞ ▼❛r❦♦✈✐❛♥✳
- ❲❡ ❤❛✈❡ ❛❧r❡❛❞② s❡❡♥ t❤❡ ✐❞❡❛ ♦❢ t❤❡ st♦❝❤❛st✐❝ ❜❛♥❞✐t
♣r♦❜❧❡♠✳ ◆♦✇ ✇❡ ❞❡✜♥❡ ✐t ♠♦r❡ ❢♦r♠❛❧❧②✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤❡ st♦❝❤❛st✐❝ ❜❛♥❞✐t ♣r♦❜❧❡♠
❙❡t✲✉♣
- ✐✈❡♥ K ≥ ✷ ❛r♠s ❛♥❞ s❡q✉❡♥❝❡s Xi,✶✱ Xi,✷✱ . . . ♦❢ ✉♥❦♥♦✇♥
r❡✇❛r❞s ❛ss♦❝✐❛t❡❞ ✇✐t❤ ❡❛❝❤ ❛r♠ i = ✶, . . . , K✱ ✇❡ st✉❞② ❢♦r❡❝❛st❡rs t❤❛t ❛t ❡❛❝❤ t✐♠❡ st❡♣ t = ✶, ✷, . . . s❡❧❡❝t ❛♥ ❛r♠ It ❛♥❞ r❡❝❡✐✈❡ t❤❡ ❛ss♦❝✐❛t❡❞ r❡✇❛r❞ XIt,t✳
❚❤❡ st♦❝❤❛st✐❝ ❜❛♥❞✐t ♣r♦❜❧❡♠
❑♥♦✇♥ ♣❛r❛♠❡t❡rs✿ ♥✉♠❜❡r ♦❢ ❛r♠s K ❛♥❞ ♦❢ r♦✉♥❞s n ≥ K✳ ❯♥❦♥♦✇♥ ♣❛r❛♠❡t❡rs✿ K ♣r♦❜❛❜✐❧✐t② ❞✐str✐❜✉t✐♦♥s ν✶, . . . , νK ♦♥ [✵, ✶]✳ ❋♦r ❡❛❝❤ r♦✉♥❞ t = ✶, ✷, . . .
- t❤❡ ❢♦r❡❝❛st❡r ❝❤♦♦s❡s It ∈ {✶, . . . , K}❀
- ❣✐✈❡♥ It✱ t❤❡ ❡♥✈✐r♦♥♠❡♥t ❞r❛✇s t❤❡ r❡✇❛r❞ XIt,t ∼ νIt
✐♥❞❡♣❡♥❞❡♥t❧② ❢r♦♠ t❤❡ ♣❛st ❛♥❞ r❡✈❡❛❧s ✐t t♦ t❤❡ ❢♦r❡❝❛st❡r✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❚❤❡ ❣♦❛❧ ✐s t❤❡ s❛♠❡ t❤❛♥ ❜❡❢♦r❡✿ ♠❛①✐♠✐③❡ t❤❡ r❡✇❛r❞✳
- ❆♥❛❧♦❣♦✉s❧② t♦ t❤✐s ✐s t♦ ♠✐♥✐♠✐③❡ t❤❡ r❡❣r❡t✳
- ❚❤❡ r❡❣r❡t ✐s ✇❤❛t ✇❡ ❧♦s❡ ❜② ♥♦t ♣❧❛②✐♥❣ t❤❡ ♦♣t✐♠❛❧ str❛t❡❣②✿
Rn = ♠❛①
i=✶,...,K n
- t=✶
Xi,t −
n
- t=✶
XIt,t. ✭✾✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❊①♣❡❝t❡❞ r❡❣r❡t✿
E[Rn] = E
- ♠❛①
i=✶,...,K n
- t=✶
Xi,t −
n
- t=✶
XIt,t
- ✭✶✵✮
- Ps❡✉❞♦✲r❡❣r❡t✿
¯ Rn = ♠❛①
i=✶,...,K E[ n
- t=✶
Xi,t −
n
- t=✶
XIt,t] ✭✶✶✮
- ◆♦t❡✿ ❚❤❡ ♣s❡✉❞♦✲r❡❣r❡t ✐s ❛ ✇❡❛❦❡r ♥♦t✐♦♥ ♦❢ r❡❣r❡t✱ s✐♥❝❡
♦♥❡ ❝♦♠♣❡t❡s ❛❣❛✐♥st t❤❡ ❛❝t✐♦♥ ✇❤✐❝❤ ✐s ♦♣t✐♠❛❧ ♦♥❧② ✐♥ ❡①♣❡❝t❛t✐♦♥✳ ▼♦r❡ ❢♦r♠❛❧❧②✿ ¯ Rn ≤ E[Rn] ✭✶✷✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ■♥ t❤❡ st♦❝❤❛st✐❝ s❡tt✐♥❣✱ t❤❡ ♣s❡✉❞♦✲r❡❣r❡t ❝❛♥ ❜❡ ✇r✐tt❡♥ ❛s
¯ Rn = n ♠❛①
i=✶,...,k µi − n
- t=✶
E[µIt] ✭✶✸✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤♦♠♣s♦♥ ❙❛♠♣❧✐♥❣
- ■♥ ♦♥❡ ♦❢ t❤❡ ❡❛r❧✐❡st ✇♦r❦s ♦♥ st♦❝❤❛st✐❝ ❜❛♥❞✐t ♣r♦❜❧❡♠s✱
❚❤♦♠♣s♦♥ ♣r♦♣♦s❡❞ ❛ r❛♥❞♦♠✐③❡❞ ❇❛②❡s✐❛♥ ❛❧❣♦r✐t❤♠ t♦ ♠✐♥✐♠✐③❡ r❡❣r❡t✳
- ❇❛s✐❝ ✐❞❡❛✿ ❛ss✉♠❡ ❛ s✐♠♣❧❡ ♣r✐♦r ❞✐str✐❜✉t✐♦♥ ♦♥ t❤❡
♣❛r❛♠❡t❡rs ♦❢ t❤❡ r❡✇❛r❞ ❞✐str✐❜✉t✐♦♥ ♦❢ ❡✈❡r② ❛r♠✱ ❛♥❞ ❛t ❛♥② t✐♠❡ st❡♣✱ ♣❧❛② ❛♥ ❛r♠ ❛❝❝♦r❞✐♥❣ t♦ ✐ts ♣♦st❡r✐♦r ♣r♦❜❛❜✐❧✐t② ♦❢ ❜❡✐♥❣ t❤❡ ❜❡st ❛r♠✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤♦♠♣s♦♥ ❙❛♠♣❧✐♥❣ ✲ ✷
- ❲❡ ❛ss✉♠❡ t❤❛t t❤❡ r❡✇❛r❞ µi ∈ [✵, ✶] ❤❛✈❡ ❛♥ ✐♥✐t✐❛❧
❞✐str✐❜✉t✐♦♥✱ ❛♥❞ πi,t ❜❡ t❤❡ ♣♦st❡r✐♦r ❞✐str✐❜✉t✐♦♥ ❢♦r µi ❛t t✐♠❡ t✳ ❚❤❡ r❡✇❛r❞ ❛t t✐♠❡ t ♦❢ ❛r♠ i ✐s θi,t ∼ πi,t✱ ✇❤❡r❡ θi,t ❛r❡ ✐♥❞❡♣❡♥❞❡♥t✳ ❚❤❡ str❛t❡❣② ✐s ✐♠♣❧❡♠❡♥t❡❞ ❜② s❛♠♣❧✐♥❣ ❢r♦♠ t❤❡ ♣♦st❡r✐♦r✳
- ❚❤❡ str❛t❡❣② ✐s t❤❡♥ ❣✐✈❡♥ ❜② ❛r❣♠❛①
i=✶,...,K
θi,t✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤♦♠♣s♦♥ ❙❛♠♣❧✐♥❣ ❢♦r t❤❡ ❇❡r♥♦✉❧❧✐ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ■♥ t❤✐s ❝❛s❡✱ t❤❡ r❡✇❛r❞s ❛r❡ ❡✐t❤❡r ✵ ♦r ✶✱ ❛♥❞ ❢♦r ❛r♠ i✱ t❤❡
♣r♦❜❛❜✐❧✐t② ♦❢ s✉❝❝❡ss ✐s µi✳
- ❇❡t❛ ❞✐str✐❜✉t✐♦♥ t✉r♥s ♦✉t t♦ ❜❡ ❛ ✈❡r② ❝♦♥✈❡♥✐❡♥t ❝❤♦✐❝❡ ♦❢
♣r✐♦rs ❢♦r ❇❡r♥♦✉❧❧✐ r❡✇❛r❞s✿
- f (x; α, β) =
Γ(α+β) Γ(α)Γ(β)xα−✶(✶ − x)β−✶
✳
- ❲❤②❄ ❇❡❝❛✉s❡ ✐❢ t❤❡ ♣r✐♦r ✐s ❛ ❇❡t❛(α, β) ❞✐str✐❜✉t✐♦♥✱ t❤❡♥
❛❢t❡r ♦❜s❡r✈✐♥❣ ❛ ❇❡r♥♦✉❧❧✐ tr✐❛❧✱t❤❡ ♣♦st❡r✐♦r ❞✐str✐❜✉t✐♦♥ ✐s s✐♠♣❧② ❇❡t❛(α + ✶, β) ♦r ❇❡t❛(α, β + ✶) ❞❡♣❡♥❞✐♥❣ ♦♥ ✇❤❡t❤❡r t❤❡ tr✐❛❧ r❡s✉❧t❡❞ ✐♥ ❛ s✉❝❝❡ss ♦r ❢❛✐❧✉r❡✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤♦♠♣s♦♥ ❙❛♠♣❧✐♥❣ ❢♦r ❇❡r♥♦✉❧❧✐ ❜❛♥❞✐t ✲ ❆❧❣♦r✐t❤♠
❚❤❡ ❚❤♦♠♣s♦♥ ❙❛♠♣❧✐♥❣ ❛❧❣♦r✐t❤♠ ✐♥✐t✐❛❧❧② ❛ss✉♠❡s ❛r♠ i t♦ ❤❛✈❡ ♣r✐♦r ❇❡t❛(✶, ✶) ♦♥ µi✱ ✇❤✐❝❤ ✐s ♥❛t✉r❛❧ ❜❡❝❛✉s❡ ❇❡t❛(✶, ✶) ✐s t❤❡ ✉♥✐❢♦r♠ ❞✐str✐❜✉t✐♦♥ ♦♥ (✵, ✶)✳ ❋♦r ❡❛❝❤ ❛r♠ i = ✶, . . . , N s❡t Si = ✵✱ Fi = ✵✳ ❋♦r ❡❛❝❤ t = ✶, ✷, . . . ❞♦✿
- ❋♦r ❡❛❝❤ ❛r♠ i = ✶, . . . , N✱ s❛♠♣❧❡ µi(t) ❢r♦♠ t❤❡ ❇❡t❛✭Si + ✶✱
Fi + ✶✮ ❞✐str✐❜✉t✐♦♥✳
- P❧❛② ❛r♠ i(t) = ❛r♠❛①µi(t)✳
- ■❢ r = ✶✱ Si(t) = Si(t) + ✶ ❡❧s❡ Fi(t) = Fi(t) + ✶
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤❡ ❛❞✈❡rs✐❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❙✉♣♣♦s❡ ✇❡ ❛r❡ ✐♥ ❛ r✐❣❣❡❞ ❝❛s✐♥♦✳
- ❚❤❡ ♦✇♥❡r✴❛❞✈❡rs❛r② s❡ts t❤❡ ❣❛✐♥ Xi,t t♦ s♦♠❡ ❛r❜✐tr❛r② ✈❛❧✉❡
gi,t ∈ [✵, ✶]✳
- ❚❤❡ ♦✇♥❡r ✐s ❝❛❧❧❡❞ ♦❜❧✐✈✐♦✉s ✐❢ ❤✐s ❝❤♦✐❝❡ ❞♦❡s ♥♦t ❞❡♣❡♥❞ ♦♥
t❤❡ str❛t❡❣② ♦❢ t❤❡ ♣❧❛②❡r✳
- ■t✬s ❝❛❧❧❡❞ ♥♦♥✲♦❜❧✐✈✐♦✉s ✐❢ ❤✐s ❝❤♦✐❝❡ ❞❡♣❡♥❞s ♦♥ t❤❡ str❛t❡❣②
♦❢ t❤❡ ♣❧❛②❡r✱ ✇❤✐❝❤ ♠❡❛♥s✿ gi,t = gi,t(I✶, . . . , It−✶) ✭✶✹✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤❡ ❛❞✈❡rs✐❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✲ ✷
❑♥♦✇♥ ♣❛r❛♠❡t❡rs✿ ♥✉♠❜❡r ♦❢ ❛r♠s K ≥ ✷ ❛♥❞ ✭♣♦ss✐❜❧②✮ ♥✉♠❜❡r ♦❢ r♦✉♥❞s n ≥ K✳ ❋♦r ❡❛❝❤ r♦✉♥❞ t = ✶, ✷, . . . ✭✶✮ t❤❡ ❢♦r❡❝❛st❡r ❝❤♦♦s❡s It ∈ {✶, . . . , K}✱ ♣♦ss✐❜❧② ✇✐t❤ t❤❡ ❤❡❧♣ ♦❢ ❡①t❡r♥❛❧ r❛♥❞♦♠✐③❛t✐♦♥✱ ✭✷✮ s✐♠✉❧t❛♥❡♦✉s❧②✱ t❤❡ ❛❞✈❡rs❛r② s❡❧❡❝ts ❛ ❣❛✐♥ ✈❡❝t♦r gt = (g✶,t, . . . , gK,t) ∈ [✵, ✶]K✱ ♣♦ss✐❜❧② ✇✐t❤ t❤❡ ❤❡❧♣ ♦❢ ❡①t❡r♥❛❧ r❛♥❞♦♠✐③❛t✐♦♥✱❛♥❞ ✭✸✮ t❤❡ ❢♦r❡❝❛st❡r r❡❝❡✐✈❡s ✭❛♥❞ ♦❜s❡r✈❡s✮ t❤❡ r❡✇❛r❞ gIt,t ✇❤✐❧❡ t❤❡ ❣❛✐♥s ♦❢ t❤❡ ♦t❤❡r ❛r♠s ❛r❡ ♥♦t ♦❜s❡r✈❡❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚❤❡ ❛❞✈❡rs✐❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✲ ✸
- ❋♦r t❤❡ ❛❞✈❡rs✐❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠✱ t❤❡ ♣s❡✉❞♦✲r❡❣r❡t ❝❛♥ ❜❡
❞❡✜♥❡❞ ❛s ¯ Rn = ♠❛①
i=✶,...,K E
n
- t=✶
gi,t −
n
- t=✶
gIt,t
- ✭✶✺✮
- ◆♦t❡✿ ❚❤❡ ❛❞✈❡rs✐❛❧ ❣❛✐♥s gi,t(I✶, . . . , It−✶) ❝♦✉❧❞ ❜❡ ❞✐✛❡r❡♥t
t❤❛♥ t❤♦s❡ ❝❤♦s❡♥ ❜② t❤❡ ♣❧❛②❡r ✇✐t❤ t❤❡ ♦♣t✐♠❛❧ t❛❝t✐❝✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▼❛r❦♦✈✐❛♥ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ■♥ t❤❡ ▼❛r❦♦✈✐❛♥ ❜❛♥❞✐t ♣r♦❜❧❡♠✱ t❤❡ r❡✇❛r❞ ♣r♦❝❡ss❡s ❛r❡
♥❡✐t❤❡r ✐✳✐✳❞✳ ♥♦r ❛❞✈❡rs✐❛❧✳
- ❆r♠s ❛r❡ ❛ss♦❝✐❛t❡❞ ✇✐t❤ K ▼❛r❦♦✈ ♣r♦❝❡ss❡s✱ ❡❛❝❤ ✇✐t❤ ✐ts
♦✇♥ st❛t❡ s♣❛❝❡✳
- ❊❛❝❤ t✐♠❡ ❛♥ ❛r♠ i ✐s ❝❤♦s❡♥ ✐♥ st❛t❡ s✱ ❛ st♦❝❤❛st✐❝ r❡✇❛r❞ ✐s
❞r❛✇♥ ❢r♦♠ ❛ ♣r♦❜❛❜✐❧✐t② ❞✐str✐❜✉t✐♦♥ νi,s✱ ❛♥❞ t❤❡ st❛t❡ ♦❢ t❤❡ r❡✇❛r❞ ♣r♦❝❡ss ❢♦r ❛r♠ i ❝❤❛♥❣❡s ✐♥ ❛ ▼❛r❦♦✈✐❛♥ ❢❛s❤✐♦♥✱ ❜❛s❡❞ ♦♥ ❛♥ ✉♥❞❡r❧②✐♥❣ st♦❝❤❛st✐❝ tr❛♥s✐t✐♦♥ ♠❛tr✐① Mi✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❈♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❙♦ ❢❛r✿ ♦♥❧② ♥♦♥ ❛ss♦❝✐❛t✐✈❡ t❛s❦s✱ ✐♥ ✇❤✐❝❤ t❤❡r❡ ✐s ♥♦ ♥❡❡❞ t♦
❛ss♦❝✐❛t❡ ❞✐✛❡r❡♥t ❛❝t✐♦♥s ✇✐t❤ ❞✐✛❡r❡♥t s✐t✉❛t✐♦♥s✳ ■♥ t❤❡s❡ t❛s❦s t❤❡ ❧❡❛r♥❡r ❡✐t❤❡r tr✐❡s t♦ ✜♥❞ ❛ s✐♥❣❧❡ ❜❡st ❛❝t✐♦♥ ✇❤❡♥ t❤❡ t❛s❦ ✐s st❛t✐♦♥❛r②✱ ♦r tr✐❡s t♦ tr❛❝❦ t❤❡ ❜❡st ❛❝t✐♦♥ ❛s ✐t ❝❤❛♥❣❡s ♦✈❡r t✐♠❡ ✇❤❡♥ t❤❡ t❛s❦ ✐s ♥♦♥ st❛t✐♦♥❛r②✳
- ◆♦✇✿ ✐♥ t❤❡ ❛ss♦❝✐❛t✐✈❡ s❡❛r❝❤✱ ✇❡ ❛❝t ❞✐✛❡r❡♥t❧② ❛❝❝♦r❞✐♥❣ t♦
t❤❡ s✐t✉❛t✐♦♥ ✇❡ ❛r❡ ✐♥✱ ❜❡❢♦r❡ t❛❦✐♥❣ ❛♥ ❛❝t✐♦♥✳
- ❊①❛♠♣❧❡✿ ②♦✉ ❛r❡ ❝♦♥❢r♦♥t❡❞ ✇✐t❤ s❡✈❡r❛❧ ♥✲❛r♠❡❞ ❜❛♥❞✐ts
❝❤♦s❡♥ ❛t r❛♥❞♦♠ ❛♥❞ ②♦✉r ♦♥❧② ❝❧✉❡ ❛❜♦✉t t❤❡♠ ✐s✱ ❢♦r ❡①❛♠♣❧❡✱ t❤❡✐r ❝♦❧♦✉r✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
◆❡✇s r❡❝♦♠♠❡♥❞❛t✐♦♥ ❡①❛♠♣❧❡
- ▲❡ts s❡❡ ❛♥ ❛♣♣❧✐❝❛t✐♦♥ ♦❢ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❲❡❜ s❡r✈✐❝❡s tr② t♦ ❛❞❛♣t t❤❡✐r s❡r✈✐❝❡s ✭❛❞✈❡rt✐s❡♠❡♥t✱ ♥❡✇s✱
❛rt✐❝❧❡s✱ ✳✳✳✮ t♦ ✐♥❞✐✈✐❞✉❛❧ ✉s❡rs ❜② ♠❛❦✐♥❣ ✉s❡ ♦❢ ✉s❡r ✐♥❢♦r♠❛t✐♦♥✳
- ❚❤✐s ❝❛♥ ❜❡ s❡❡♥ ❛s ❛ ♠✉❧t✐✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✇❤❡r❡ ✇❡
❤❛✈❡ ❛❞❞✐t✐♦♥❛❧ ✐♥❢♦r♠❛t✐♦♥ ❛❜♦✉t t❤❡ s❧♦t ♠❛❝❤✐♥❡✳
- ❖♥❡ ❝❧✐❝❦ ♦❢ t❤❡ ✉s❡r ♦♥ t❤❡ ❛❞✈❡rt✐s❡♠❡♥t ♦r ♥❡✇ ❝♦rr❡s♣♦♥❞
t♦ t❤❡ r❡✇❛r❞✳
- ❚❤❡ ❛❝t✐♦♥ ❝♦rr❡s♣♦♥❞s t♦ t❤❡ ❝❤♦✐❝❡ ♦❢ t❤❡ ❛❞✈❡rt✐s❡♠❡♥t ♦r
♥❡✇ ✇❤✐❝❤ ✇❡ t❤✐♥❦ t❤❡ ✉s❡r ✇✐❧❧ ♠♦st ❧✐❦❡❧② ❝❧✐❝❦ ✭s♦ ❢♦r ✇❤✐❝❤ ✇❡ ✇♦✉❧❞ ♠♦st ❧✐❦❡❧② ❣♦t t❤❡ r❡✇❛r❞✮✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❊①❛♠♣❧❡ ♦❢ ❛ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❲❡ ❤❛✈❡ t❤❡ ❝❤♦✐❝❡ ❜❡t✇❡❡♥ ❢♦✉r ❧✐♥❦s ❛♥❞ ✇❡ ❤❛✈❡ t♦ ❞❡❝✐❞❡
✇❤✐❝❤ ♦♥❡ ✇❡ ❤✐❣❤❧✐❣❤t ♦♥ t❤❡ ②❛❤♦♦ ❢r♦♥t ♣❛❣❡✱ ✇✐t❤ t❤❡ ❛✐♠ t❤❛t t❤❡ ✉s❡r ❝❧✐❝❦s ♦♥ ✐t✳ ❲❡ ❤❛✈❡ ❛❧s♦ ✐♥❢♦r♠❛t✐♦♥ ❛❜♦✉t t❤❡ ✉s❡r✱ ❛♥❞ t❤❡r❡❢♦r❡ ❤✐❣❤❧✐❣❤t ❛♥ ❛rt✐❝❧❡ ❛❝❝♦r❞✐♥❣ t♦ ❤✐s ✐♥t❡r❡sts✳
- ❚❤✐s ♣r♦❜❧❡♠ ❝❛♥ ❜❡ s❡❡♥ ❛s ❛ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠✱
✇❤❡r❡ t❤❡ r❡✇❛r❞ ✐s ❛ ❝❧✐❝❦ ♦♥ t❤❡ ❧✐♥❦ ✇❡ ❤❛✈❡ ❤✐❣❤❧✐❣❤t❡❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❈♦♥t❡①t✉❛❧ ❜❛♥❞✐t ✲ ❛❧❣♦r✐t❤♠
- ❆ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ❛❧❣♦r✐t❤♠ A ♣r♦❝❡❡❞s ✐♥ ❞✐s❝r❡t❡ tr✐❛❧s
t = ✶, ✷, ✸, . . . ✳ ■♥ tr✐❛❧ t✿ ✶✳ ❚❤❡ ❛❧❣♦r✐t❤♠ ♦❜s❡r✈❡s t❤❡ ❝✉rr❡♥t ✉s❡r ut ❛♥❞ ❛ s❡t At ♦❢ ❛r♠s ♦r ❛❝t✐♦♥s t♦❣❡t❤❡r ✇✐t❤ t❤❡✐r ❢❡❛t✉r❡ ✈❡❝t♦rs xt,a ❢♦r a ∈ A✳ ❚❤❡ ✈❡❝t♦r xt,a s✉♠♠❛r✐③❡s ✐♥❢♦r♠❛t✐♦♥ ♦❢ ❜♦t❤ t❤❡ ✉s❡r ut ❛♥❞ ❛r♠ a✱ ❛♥❞ ✇✐❧❧ ❜❡ r❡❢❡rr❡❞ ❛s t❤❡ ❝♦♥t❡①t✳ ✷✳ ❇❛s❡❞ ♦♥ ♦❜s❡r✈❡❞ ♣❛②♦✛s ✐♥ ♣r❡✈✐♦✉s tr✐❛❧s✱ A ❝❤♦♦s❡s ❛♥ ❛r♠ at ∈ At✱ ❛♥❞ r❡❝❡✐✈❡s ♣❛②♦✛ rt,at ✇❤♦s❡ ❡①♣❡❝t❛t✐♦♥ ❞❡♣❡♥❞s ♦♥ ❜♦t❤ t❤❡ ✉s❡r ut ❛♥❞ t❤❡ ❛r♠ at✳ ✸✳ ❚❤❡ ❛❧❣♦r✐t❤♠ t❤❡♥ ✐♠♣r♦✈❡s ✐ts ❛r♠✲s❡❧❡❝t✐♦♥ str❛t❡❣② ✇✐t❤ t❤❡ ♥❡✇ ♦❜s❡r✈❛t✐♦♥ (xt,a, at, rt,at)✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❚✲tr✐❛❧ r❡❣r❡t RA
- ❚❤❡ t♦t❛❧ T✲tr✐❛❧ ♣❛②♦✛ ♦❢ A ✐s ❞❡✜♥❡❞ ❛s T
t=✶ rt,at✳
- ❉❡✜♥❡ t❤❡ ♦♣t✐♠❛❧ ❡①♣❡❝t❡❞ T✲tr✐❛❧ ♣❛②♦✛ ❛s E[T
t=✶ rt,a∗
t ]✱
✇❤❡r❡ a∗
t ✐s t❤❡ ❛r♠ ✇✐t❤ ♠❛①✐♠✉♠ ❡①♣❡❝t❡❞ ♣❛②♦✛ ❛t tr✐❛❧ t✳
- ●♦❛❧✿ ✇r✐t❡ ❛♥ ❛❧❣♦r✐t❤♠ A s♦ t❤❛t t❤❡ ❡①♣❡❝t❡❞ ♣❛②♦✛ ✐s
♠❛①✐♠✐③❡❞✳
- ❆s ✇❡ ❤❛✈❡ ❛❧r❡❛❞② s❡❡♥ t❤✐s ✐s ❡q✉✐✈❛❧❡♥t t♦ ✜♥❞ ❛♥ ❛❧❣♦r✐t❤♠
✇❤✐❝❤ ♠✐♥✐♠✐③❡s t❤❡ r❡❣r❡t ✇✐t❤ r❡s♣❡❝t t♦ t❤❡ ♦♣t✐♠❛❧ ❛r♠✲s❡❧❡❝t✐♦♥ str❛t❡❣②✿ RA(T) = E[
T
- t=✶
rt,a∗
t ] − E[
T
- t=✶
rt,at]. ✭✶✻✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆ s♣❡❝✐❛❧ ❝❛s❡ ♦❢ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
❆ s♣❡❝✐❛❧ ❝❛s❡ ♦❢ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✐s t❤❡ n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✱ ✐♥ ✇❤✐❝❤✿
- ❚❤❡ ❛r♠ s❡t A r❡♠❛✐♥s ✉♥❝❤❛♥❣❡❞ ❛♥❞ ❝♦♥t❛✐♥s n ❛r♠s ∀t✳
- ❚❤❡ s✐t✉❛t✐♦♥ st ✐s t❤❡ s❛♠❡ ∀t✳
❚❤❡ n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✐s ❛❧s♦ ❝❛❧❧❡❞ t❤❡ ❝♦♥t❡①t ❢r❡❡ ❜❛♥❞✐t ♣r♦❜❧❡♠✳ ❲❤✐❧❡ t❤❡ ❝♦♥t❡①t ❢r❡❡ ❜❛♥❞✐t ♣r♦❜❧❡♠s ❛r❡ ❡①t❡♥s✐✈❡❧② st✉❞✐❡❞ ❛♥❞ ✇❡❧❧ ✉♥❞❡rst♦♦❞✱ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ❤❛s r❡♠❛✐♥❡❞ ❝❤❛❧❧❡♥❣✐♥❣✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
- ❲❡ ✇❛♥t t♦ ✜♥❞ ❛♥ ❛❧❣♦r✐t❤♠ t❤❛t s♦❧✈❡s t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t
♣r♦❜❧❡♠✳
- ■❞❡❛✿ ✉s❡ t❤❡ ❯❈❇ ♠❡t❤♦❞
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▲✐♥❡❛r ❯♣♣❡r ❇♦✉♥❞ ❈♦♥✜❞❡♥❝❡ ✲ ✶
- ❲❡ ❛ss✉♠❡ t❤❡ ❡①♣❡❝t❡❞ ♣❛②♦✛ ♦❢ ❛♥ ❛r♠ a ✐s ❧✐♥❡❛r ✐♥ ✐ts
d✲❞✐♠❡♥s✐♦♥❛❧ ❢❡❛t✉r❡ xt,a ✇✐t❤ s♦♠❡ ✉♥❦♥♦✇♥ ❝♦❡✣❝✐❡♥t ✈❡❝t♦r θ∗
a✱ ♥❛♠❡❧②✱ ∀t✱
rt,a = xT
t,aθ∗ a + ǫt
✭✶✼✮ ✇❤✐❝❤ ✐s ❡q✉✐✈❛❧❡♥t t♦ E[rt,a | xt,a] = xT
t,aθ∗ a✳ ❚❤❡ ♣❛r❛♠❡t❡rs
❛r❡ ♥♦t s❤❛r❡❞ ❛♠♦♥❣ ❞✐✛❡r❡♥t ❛r♠s✳
- ❲❡ ✇✐s❤ t♦ ♠✐♥✐♠✐③❡ t❤❡ sq✉❛r❡ ❧♦ss
ˆ θa = ❛r❣♠✐♥
θ∗
a
m
- t=✶
(rt,a − xT
t,aθ∗ a)✷
✭✶✽✮
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▲✐♥❡❛r ❯♣♣❡r ❇♦✉♥❞ ❈♦♥✜❞❡♥❝❡ ✲ ✷
- Da ❞❡s✐❣♥ ♠❛tr✐① ♦❢ ❞✐♠❡♥s✐♦♥ m × d ❛t tr✐❛❧ t✱ ba ∈ Rm
r❡s♣♦♥s❡ ✈❡❝t♦r✳
- ❆♣♣❧②✐♥❣ r✐❞❣❡ r❡❣r❡ss✐♦♥ t♦ t❤❡ tr❛✐♥✐♥❣ ❞❛t❛ (Da, ba) ❣✐✈❡s
ˆ θa = (DT
a Da + Ia)−✶DT a ba✱ ✇❤❡r❡ Id ✐s t❤❡ d × d ✐❞❡♥t✐t②
♠❛tr✐①✳
- ❚❤❡♦r❡♠✿ ❢♦r t❤❡ ❡st✐♠❛t❡❞ ❝♦❡✣❝✐❡♥ts ✐t ❤♦❧❞s t❤❛t
| xT
t,aˆ
θa − E[rt,a | xt,a] |≤ α
- xT
t,a(DT a Da + Id)−✶xt,a
✭✶✾✮ ❢♦r ❛♥② δ > ✵ ❛♥❞ xt,a ∈ Rd✱ ✇❤❡r❡ α = ✶ +
- ❧♥(✷/δ)
✷
✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▲✐♥❡❛r ❯♣♣❡r ❇♦✉♥❞ ❈♦♥✜❞❡♥❝❡ ✲ ✸
- ❚❤❡ ✐♥❡q✉❛❧✐t② ❣✐✈❡s ❛ r❡❛s♦♥❛❜❧② t✐❣❤t ❯❈❇ ❢♦r t❤❡ ❡①♣❡❝t❡❞
♣❛②♦✛ ♦❢ ❛r♠✳ ❆t ❡❛❝❤ tr✐❛❧ t✱ ❝❤♦♦s❡✿ at = ❛r❣♠❛①
a∈At
(xT
t,aˆ
θa + α
- xT
t,aA−✶ a xt,a),
✭✷✵✮ ✇❤❡r❡ Aa = DT
a Da + Id✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▲✐♥❯❈❇ ❛❧❣♦r✐t❤♠
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❙✉♠♠❛r②
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❙❡tt✐♥❣✿ ✇❡ ❤❛✈❡ t✇♦ ❝❤♦✐❝❡s
- ■♥ t❤❡ ✜rst ♦♥❡ ✇❡ ❤❛✈❡ t♦ ❞❡❝✐❞❡ ❜❡t✇❡❡♥ n ❛❝t✐♦♥s✱ ❡✳❣✳ ❤✐t✱
st❛♥❞ ♦r ❞♦✉❜❧❡ ❞♦✇♥✳ ❲❡ ✐❣♥♦r❡ s♣❧✐tt✐♥❣✳
- ■♥ t❤❡ s❡❝♦♥❞ ♦♥❡ ✇❡ ❛❧✇❛②s st❛♥❞✳
❚❤❡ ✜rst ❞❡❝✐s✐♦♥ ❝❛♥ ❜❡ s❡❡♥ ❛s ❛ n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✳ ❖✉r ❡①❛♠♣❧❡ ❝❛♥ ❜❡ s❡❡♥ ❛s ❛ ✸✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✱ s✐♥❝❡ ✇❡ ❤❛✈❡ t❤❡ ❝❤♦✐❝❡ ❜❡t✇❡❡♥ ✸ ❛❝t✐♦♥s ✭❤✐t✱ st❛♥❞ ❛♥❞ ❞♦✉❜❧❡ ❞♦✇♥✮✳ ❚❤❡② ❤❛✈❡ ❛ ❞✐✛❡r❡♥t r❡✇❛r❞ ❞✐str✐❜✉t✐♦♥ ❛♥❞ s♦ ❛ ❞✐✛❡r❡♥t r❡✇❛r❞ ❡①♣❡❝t❛t✐♦♥✳
- ♦❛❧✿ ▼❛①✐♠✐③❡ t❤❡ r❡✇❛r❞✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦ ✲ ❣r❡❡❞② ♠❡t❤♦❞s
❲❡ ❤❛✈❡ s❡❡♥ s♦♠❡ ♠❡t❤♦❞s t♦ s♦❧✈❡ t❤❡ n✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✱ ❧❡t✬s ❛♣♣❧② t❤❡♠ t♦ ♦✉r ✸✲❛r♠❡❞ ❜❛♥❞✐t ♣r♦❜❧❡♠✳
- ❚❤❡ ❣r❡❡❞② ♠❡t❤♦❞✿ ❝❤♦♦s❡ ❛t r❛♥❞♦♠ ❡✳❣✳ t = ✶✵✵✵ t✐♠❡s
♦♥❡ ♦❢ t❤✐s ✸ ❛❝t✐♦♥s ❛♥❞ t❤❡♥ ♠❛❦❡ ②♦✉r ❡st✐♠❛t♦r ❢♦r t❤❡ ❡①♣❡❝t❛t✐♦♥ ♦❢ t❤❡ r❡✇❛r❞✿ Q(a) =
R✶+···+RNt (a) Nt(a)
N(a) = t✐♠❡s ✇❡ ❤❛✈❡ ❝❤♦s❡♥ t❤❡ ❛❝t✐♦♥ a ✉♥t✐❧ t✐♠❡ t✳ ❚❤❡♥ ❡①♣❧♦✐t t❤❡ ❣r❡❡❞② ❛❝t✐♦♥s✱ ✐✳❡✳ ❝❤♦♦s❡ ❛❧✇❛②s t❤❡ ❛❝t✐♦♥ ✇✐t❤ t❤❡ ❣r❡❛t❡st ❡st✐♠❛t♦r ❡①♣❡❝t❛t✐♦♥✳
- ❚❤❡ ǫ ❣r❡❡❞② ♠❡t❤♦❞✿ ❝❤♦♦s❡ ✇✐t❤ ✇✐t❤ ♣r♦❜❛❜✐❧✐t② ǫ ❛♥ ❛❝t✐♦♥
✇❤✐❝❤ ✐s ♥♦t t❤❡ ❣r❡❡❞② ❛❝t✐♦♥✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦ ✲ ♦t❤❡r ♠❡t❤♦❞s
- ■♥ t❤❡ ❯♣♣❡r✲❈♦♥✜❞❡♥❝❡✲❇♦✉♥❞ ❆❝t✐♦♥ ❙❡❧❡❝t✐♦♥ ▼❡t❤♦❞ ✇❡
✉s❡ t❤❡ ♥✐❝❡ ❢♦r♠✉❧❛✿ At = ❛r❣♠❛①
a
[Qt(a) + c
- ❧♥ t
Nt(a)]✳
- ❚❤❡ ❚❤♦♠♣s♦♥ s❛♠♣❧✐♥❣ ♠❡t❤♦❞s ✐❢ ✇❡ ❛❞❛♣t t❤❡ r❡✇❛r❞ s✉❝❤
t❤❛t t❤❡② ❛r❡ ❜♦✉♥❞❡❞ ✐♥ [✵, ✶]✱ ❡t❝✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦ ✲ ❈♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠
- ❚❤❡r❡ ✐s ❛♥♦t❤❡r ✇❛② t♦ ❛♣♣r♦❛❝❤ t❤✐s ♣r♦❜❧❡♠✿ ✇❡ ❝❛♥ s❡❡ ✐t
❛s ❛ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠✳
- ❚❤❡ ✐❞❡❛ ✐s t♦ ❧♦♦❦ ❛t ❡✈❡r② ♣♦ss✐❜❧❡ ❝♦♥✜❣✉r❛t✐♦♥ ♦❢ ❝❛r❞s t❤❡
❞❡❛❧❡r ❛♥❞ t❤❡ ♣❧❛②❡r ❤❛✈❡ ❛♥❞ ✇❤❛t ❛❝t✐♦♥s ✇❡r❡ t❤❡ ❜❡st ✐♥ t❤❡ ❞✐✛❡r❡♥t s✐t✉❛t✐♦♥s✱ ✐✳❡✳ ✇❡ ✐♠♣❧❡♠❡♥t ✇✐t❤ ❧♦ts ♦❢ s✐♥❣❧❡ ❜❛♥❞✐ts✱ ✇✐t❤♦✉t ✐♥❢♦r♠❛t✐♦♥ s❤❛r✐♥❣✳
- ❙♦ ❛s ✐♥ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✇❡ ❤❛✈❡ t♦ ❛ss♦❝✐❛t❡
❞✐✛❡r❡♥t ❛❝t✐♦♥s ✇✐t❤ ❞✐✛❡r❡♥t s✐t✉❛t✐♦♥s ✭❝♦♥✜❣✉r❛t✐♦♥ ♦❢ ❝❛r❞s✮✳
▼✉❧t✐✲❛r♠ ❇❛♥❞✐ts ❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦
❆♣♣❧✐❝❛t✐♦♥ t♦ ❇❧❛❝❦❥❛❝❦ ✲ ❯❈❇
- ❆ ✇❛② t♦ s♦❧✈❡ t❤❡ ❝♦♥t❡①t✉❛❧ ❜❛♥❞✐t ♣r♦❜❧❡♠ ✐s t♦ ✉s❡ t❤❡
❛❧r❡❛❞② s❡❡♥ ❧✐♥❡❛r ❯❈❇ ❛❧❣♦r✐t❤♠✿
- ❋❡❛t✉r❡ ✈❡❝t♦r xt,a ✐s t❤❡ ✐♥❢♦r♠❛t✐♦♥ ✇❡ ❤❛✈❡ ♦♥ t❤❡ ♣❧❛②❡rs
❛♥❞ ❞❡❛❧❡rs ❝❛r❞s✳
- E[rt,a | xt,a] ✐s t❤❡ ❡①♣❡❝t❛t✐♦♥ ♦❢ t❤❡ r❡✇❛r❞ ❣✐✈❡♥ xt,a✳ ●♦❛❧✿
♠❛①✐♠✐③❡ ✐t✳
- ❚❤❡ r♦✇s ♦❢ t❤❡ ❞❡s✐❣♥ ♠❛tr✐① Da ❝♦rr❡s♣♦♥❞ t♦ t❤❡ ❞✐✛❡r❡♥t