Addition of two L-smooth function is also L-smooth?
Clash Royale CLAN TAG#URR8PPP
up vote
0
down vote
favorite
Assume $f(x)$ has an L-Lipschitz continuous gradient say $L_1$ i.e there is a constant L>0 such that
$$|nabla f(x) - nabla f(y)|_2 le L|x-y|_2$$ for any $x,y$.
Also $g(x)$ has an L-Lipschitz continuous gradient, say $L_2$. Is $f(x)+g(x)$ has an L-Lipschitz continuous gradient?
I tried to use property of L2 norm but couldn't be sure if this is correct. Since
$| x + y|_2 leq | x|_2 + | y|_2$, the $L$ of $f(x)+g(x)$ would be $L_1+L_2$
I would appreciate any help.
lipschitz-functions gradient-descent non-convex-optimization
add a comment |Â
up vote
0
down vote
favorite
Assume $f(x)$ has an L-Lipschitz continuous gradient say $L_1$ i.e there is a constant L>0 such that
$$|nabla f(x) - nabla f(y)|_2 le L|x-y|_2$$ for any $x,y$.
Also $g(x)$ has an L-Lipschitz continuous gradient, say $L_2$. Is $f(x)+g(x)$ has an L-Lipschitz continuous gradient?
I tried to use property of L2 norm but couldn't be sure if this is correct. Since
$| x + y|_2 leq | x|_2 + | y|_2$, the $L$ of $f(x)+g(x)$ would be $L_1+L_2$
I would appreciate any help.
lipschitz-functions gradient-descent non-convex-optimization
add a comment |Â
up vote
0
down vote
favorite
up vote
0
down vote
favorite
Assume $f(x)$ has an L-Lipschitz continuous gradient say $L_1$ i.e there is a constant L>0 such that
$$|nabla f(x) - nabla f(y)|_2 le L|x-y|_2$$ for any $x,y$.
Also $g(x)$ has an L-Lipschitz continuous gradient, say $L_2$. Is $f(x)+g(x)$ has an L-Lipschitz continuous gradient?
I tried to use property of L2 norm but couldn't be sure if this is correct. Since
$| x + y|_2 leq | x|_2 + | y|_2$, the $L$ of $f(x)+g(x)$ would be $L_1+L_2$
I would appreciate any help.
lipschitz-functions gradient-descent non-convex-optimization
Assume $f(x)$ has an L-Lipschitz continuous gradient say $L_1$ i.e there is a constant L>0 such that
$$|nabla f(x) - nabla f(y)|_2 le L|x-y|_2$$ for any $x,y$.
Also $g(x)$ has an L-Lipschitz continuous gradient, say $L_2$. Is $f(x)+g(x)$ has an L-Lipschitz continuous gradient?
I tried to use property of L2 norm but couldn't be sure if this is correct. Since
$| x + y|_2 leq | x|_2 + | y|_2$, the $L$ of $f(x)+g(x)$ would be $L_1+L_2$
I would appreciate any help.
lipschitz-functions gradient-descent non-convex-optimization
edited Aug 20 at 14:07
Martin Sleziak
43.5k6113260
43.5k6113260
asked Aug 20 at 12:56
Pumpkin
4311417
4311417
add a comment |Â
add a comment |Â
1 Answer
1
active
oldest
votes
up vote
0
down vote
I may be wrong but this seems like a simple use of sub-additivity of the norm and additivity of the gradient:
$||nabla big( f+g big)(x) - nabla big(f+g)(y)||_2=BigVert nabla f(x)+nabla g(x)- big( nabla f(y)+nabla g(y) big) Big Vert_2=$
$=Big Vert big( nabla f(x)-nabla f(y) big) + big( nabla g(x)-nabla g(y) big) Big Vert_2leq big Vert nabla f(x)-nabla f(y) big Vert_2+ big Vert nabla g(x)-nabla g(y) big Vert_2leq$
$leq L_1cdot Vert x-y Vert_2+ L_2 cdot Vert x-y Vert_2= big( L_1+L_2 big) cdot Vert x-y Vert_2 $
add a comment |Â
1 Answer
1
active
oldest
votes
1 Answer
1
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
0
down vote
I may be wrong but this seems like a simple use of sub-additivity of the norm and additivity of the gradient:
$||nabla big( f+g big)(x) - nabla big(f+g)(y)||_2=BigVert nabla f(x)+nabla g(x)- big( nabla f(y)+nabla g(y) big) Big Vert_2=$
$=Big Vert big( nabla f(x)-nabla f(y) big) + big( nabla g(x)-nabla g(y) big) Big Vert_2leq big Vert nabla f(x)-nabla f(y) big Vert_2+ big Vert nabla g(x)-nabla g(y) big Vert_2leq$
$leq L_1cdot Vert x-y Vert_2+ L_2 cdot Vert x-y Vert_2= big( L_1+L_2 big) cdot Vert x-y Vert_2 $
add a comment |Â
up vote
0
down vote
I may be wrong but this seems like a simple use of sub-additivity of the norm and additivity of the gradient:
$||nabla big( f+g big)(x) - nabla big(f+g)(y)||_2=BigVert nabla f(x)+nabla g(x)- big( nabla f(y)+nabla g(y) big) Big Vert_2=$
$=Big Vert big( nabla f(x)-nabla f(y) big) + big( nabla g(x)-nabla g(y) big) Big Vert_2leq big Vert nabla f(x)-nabla f(y) big Vert_2+ big Vert nabla g(x)-nabla g(y) big Vert_2leq$
$leq L_1cdot Vert x-y Vert_2+ L_2 cdot Vert x-y Vert_2= big( L_1+L_2 big) cdot Vert x-y Vert_2 $
add a comment |Â
up vote
0
down vote
up vote
0
down vote
I may be wrong but this seems like a simple use of sub-additivity of the norm and additivity of the gradient:
$||nabla big( f+g big)(x) - nabla big(f+g)(y)||_2=BigVert nabla f(x)+nabla g(x)- big( nabla f(y)+nabla g(y) big) Big Vert_2=$
$=Big Vert big( nabla f(x)-nabla f(y) big) + big( nabla g(x)-nabla g(y) big) Big Vert_2leq big Vert nabla f(x)-nabla f(y) big Vert_2+ big Vert nabla g(x)-nabla g(y) big Vert_2leq$
$leq L_1cdot Vert x-y Vert_2+ L_2 cdot Vert x-y Vert_2= big( L_1+L_2 big) cdot Vert x-y Vert_2 $
I may be wrong but this seems like a simple use of sub-additivity of the norm and additivity of the gradient:
$||nabla big( f+g big)(x) - nabla big(f+g)(y)||_2=BigVert nabla f(x)+nabla g(x)- big( nabla f(y)+nabla g(y) big) Big Vert_2=$
$=Big Vert big( nabla f(x)-nabla f(y) big) + big( nabla g(x)-nabla g(y) big) Big Vert_2leq big Vert nabla f(x)-nabla f(y) big Vert_2+ big Vert nabla g(x)-nabla g(y) big Vert_2leq$
$leq L_1cdot Vert x-y Vert_2+ L_2 cdot Vert x-y Vert_2= big( L_1+L_2 big) cdot Vert x-y Vert_2 $
answered Aug 20 at 13:19
Keen-ameteur
726213
726213
add a comment |Â
add a comment |Â
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2888753%2faddition-of-two-l-smooth-function-is-also-l-smooth%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password