How to show $(u^{3/2}+v^{3/2})^{2/3} leq u+v$ where $u,v geq 0$? [closed]












1












$begingroup$


In Convex Optimization book written by Prof. Boyd on page 499 section 9.6.2 says we have



$$(u^{3/2}+v^{3/2})^{2/3} leq u+v$$



where $u,v geq 0$. How can we show that?










share|cite|improve this question









$endgroup$



closed as off-topic by Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen Jan 22 at 10:59


This question appears to be off-topic. The users who voted to close gave this specific reason:


  • "This question is missing context or other details: Please provide additional context, which ideally explains why the question is relevant to you and our community. Some forms of context include: background and motivation, relevant definitions, source, possible strategies, your current progress, why the question is interesting or important, etc." – Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen

If this question can be reworded to fit the rules in the help center, please edit the question.





















    1












    $begingroup$


    In Convex Optimization book written by Prof. Boyd on page 499 section 9.6.2 says we have



    $$(u^{3/2}+v^{3/2})^{2/3} leq u+v$$



    where $u,v geq 0$. How can we show that?










    share|cite|improve this question









    $endgroup$



    closed as off-topic by Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen Jan 22 at 10:59


    This question appears to be off-topic. The users who voted to close gave this specific reason:


    • "This question is missing context or other details: Please provide additional context, which ideally explains why the question is relevant to you and our community. Some forms of context include: background and motivation, relevant definitions, source, possible strategies, your current progress, why the question is interesting or important, etc." – Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen

    If this question can be reworded to fit the rules in the help center, please edit the question.



















      1












      1








      1





      $begingroup$


      In Convex Optimization book written by Prof. Boyd on page 499 section 9.6.2 says we have



      $$(u^{3/2}+v^{3/2})^{2/3} leq u+v$$



      where $u,v geq 0$. How can we show that?










      share|cite|improve this question









      $endgroup$




      In Convex Optimization book written by Prof. Boyd on page 499 section 9.6.2 says we have



      $$(u^{3/2}+v^{3/2})^{2/3} leq u+v$$



      where $u,v geq 0$. How can we show that?







      linear-algebra inequality






      share|cite|improve this question













      share|cite|improve this question











      share|cite|improve this question




      share|cite|improve this question










      asked Jan 14 at 21:14









      SaeedSaeed

      1,124310




      1,124310




      closed as off-topic by Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen Jan 22 at 10:59


      This question appears to be off-topic. The users who voted to close gave this specific reason:


      • "This question is missing context or other details: Please provide additional context, which ideally explains why the question is relevant to you and our community. Some forms of context include: background and motivation, relevant definitions, source, possible strategies, your current progress, why the question is interesting or important, etc." – Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen

      If this question can be reworded to fit the rules in the help center, please edit the question.







      closed as off-topic by Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen Jan 22 at 10:59


      This question appears to be off-topic. The users who voted to close gave this specific reason:


      • "This question is missing context or other details: Please provide additional context, which ideally explains why the question is relevant to you and our community. Some forms of context include: background and motivation, relevant definitions, source, possible strategies, your current progress, why the question is interesting or important, etc." – Namaste, onurcanbektas, Lee David Chung Lin, Gibbs, Kemono Chen

      If this question can be reworded to fit the rules in the help center, please edit the question.






















          5 Answers
          5






          active

          oldest

          votes


















          5












          $begingroup$

          In general if $p > 1$ and $u,v ge 0$ then $u^p + v^p le (u+v)^p$. This is evident if $u = 0$, and if $u > 0$ you can divide by $u^p$ to arrive at the equivalent inequality
          $$1 + left( frac vu right)^p le left( 1 + frac vu right)^p.$$



          The function $phi(t) = (1+t)^p - t^p - 1$, $t ge 0$, satisfies $phi'(t) = p(t+1)^{p-1} - pt^{p-1} > 0$ for $t > 0$ because $p-1 > 0$. Thus $phi$ is increasing on $[0,infty)$ so that $phi(t) ge phi(0) = 0$. That is,
          $$1 + t^p le (1+t)^p$$ for all $t ge 0$ and in particular for $t = dfrac vu$.






          share|cite|improve this answer









          $endgroup$













          • $begingroup$
            That's exactly how I was going to do it until I saw that you beat me to it.
            $endgroup$
            – marty cohen
            Jan 14 at 21:27










          • $begingroup$
            I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
            $endgroup$
            – Umberto P.
            Jan 14 at 21:30












          • $begingroup$
            @Umberto P.: Your answer is very elegant. Thank you for sharing that.
            $endgroup$
            – Saeed
            Jan 15 at 17:24










          • $begingroup$
            @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
            $endgroup$
            – Saeed
            Jan 15 at 17:28










          • $begingroup$
            @ Umberto P: Could you help me to prove it?
            $endgroup$
            – Saeed
            Jan 16 at 3:17



















          3












          $begingroup$

          It is also a consequence of Minkowski's inequality for $p=frac{3}{2}$. Let
          $$
          stackrel{rightarrow}{a}=(u,0),quadstackrel{rightarrow}{b}=(0,v).
          $$
          Then Minkowski's inequality says
          $$
          left(u^{frac{3}{2}}+v^{{frac{3}{2}}}right)^{frac{2}{3}}=|stackrel{rightarrow}{a}+stackrel{rightarrow}{b}|_{frac{3}{2}}le |stackrel{rightarrow}{a}|_{frac{3}{2}}+|stackrel{rightarrow}{b}|_{frac{3}{2}}=u+v.
          $$






          share|cite|improve this answer









          $endgroup$













          • $begingroup$
            I was wondering how we can show it with one famous inequality. Then you did it.
            $endgroup$
            – Saeed
            Jan 14 at 21:43










          • $begingroup$
            Yeah, I had the same mind when I started thinking about it :)
            $endgroup$
            – Song
            Jan 14 at 21:54



















          1












          $begingroup$

          It's
          $$u^3+3u^2v+3uv^2+v^3geq u^3+2sqrt{u^3v^3}+v^3$$ or
          $$3u+3vgeq2sqrt{uv}$$ or
          $$(sqrt{u}-sqrt{v})^2+2u+2vgeq0,$$ which is obvious.






          share|cite|improve this answer









          $endgroup$





















            1












            $begingroup$

            Write $a= sqrt{u}$ and $b= sqrt{v}$, so you have to prove:



            $$ (a^3+b^3)^2leq (a^2+b^2)^3$$
            so $$ 2a^3b^3leq 3a^4b^2+3a^2b^4$$



            so $$2ableq 3a^2+3b^2$$



            But this is true since $$2ableq a^2+b^2leq 3a^2+3b^2$$






            share|cite|improve this answer









            $endgroup$









            • 1




              $begingroup$
              $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
              $endgroup$
              – marty cohen
              Jan 14 at 21:26



















            0












            $begingroup$

            I would raise each side to the $3^{rd}$ power and expand:



            $$u^3+2(uv)^{frac 32} + v^3 le u^3 + 3u^2v + 3uv^2 + v^3$$



            Which simplifies to



            $$2(uv)^{frac 32} le 3uv(u+v)$$



            Which is



            $$2sqrt{uv} le 3u+3v$$



            And if you are familiar with the famous inequality $a^2+b^2 ge 2ab$ (which is gotten from $(a-b)^2 ge 0$) then you see that $2sqrt{uv} le 3u+3v$ as well.






            share|cite|improve this answer









            $endgroup$




















              5 Answers
              5






              active

              oldest

              votes








              5 Answers
              5






              active

              oldest

              votes









              active

              oldest

              votes






              active

              oldest

              votes









              5












              $begingroup$

              In general if $p > 1$ and $u,v ge 0$ then $u^p + v^p le (u+v)^p$. This is evident if $u = 0$, and if $u > 0$ you can divide by $u^p$ to arrive at the equivalent inequality
              $$1 + left( frac vu right)^p le left( 1 + frac vu right)^p.$$



              The function $phi(t) = (1+t)^p - t^p - 1$, $t ge 0$, satisfies $phi'(t) = p(t+1)^{p-1} - pt^{p-1} > 0$ for $t > 0$ because $p-1 > 0$. Thus $phi$ is increasing on $[0,infty)$ so that $phi(t) ge phi(0) = 0$. That is,
              $$1 + t^p le (1+t)^p$$ for all $t ge 0$ and in particular for $t = dfrac vu$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                That's exactly how I was going to do it until I saw that you beat me to it.
                $endgroup$
                – marty cohen
                Jan 14 at 21:27










              • $begingroup$
                I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
                $endgroup$
                – Umberto P.
                Jan 14 at 21:30












              • $begingroup$
                @Umberto P.: Your answer is very elegant. Thank you for sharing that.
                $endgroup$
                – Saeed
                Jan 15 at 17:24










              • $begingroup$
                @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
                $endgroup$
                – Saeed
                Jan 15 at 17:28










              • $begingroup$
                @ Umberto P: Could you help me to prove it?
                $endgroup$
                – Saeed
                Jan 16 at 3:17
















              5












              $begingroup$

              In general if $p > 1$ and $u,v ge 0$ then $u^p + v^p le (u+v)^p$. This is evident if $u = 0$, and if $u > 0$ you can divide by $u^p$ to arrive at the equivalent inequality
              $$1 + left( frac vu right)^p le left( 1 + frac vu right)^p.$$



              The function $phi(t) = (1+t)^p - t^p - 1$, $t ge 0$, satisfies $phi'(t) = p(t+1)^{p-1} - pt^{p-1} > 0$ for $t > 0$ because $p-1 > 0$. Thus $phi$ is increasing on $[0,infty)$ so that $phi(t) ge phi(0) = 0$. That is,
              $$1 + t^p le (1+t)^p$$ for all $t ge 0$ and in particular for $t = dfrac vu$.






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                That's exactly how I was going to do it until I saw that you beat me to it.
                $endgroup$
                – marty cohen
                Jan 14 at 21:27










              • $begingroup$
                I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
                $endgroup$
                – Umberto P.
                Jan 14 at 21:30












              • $begingroup$
                @Umberto P.: Your answer is very elegant. Thank you for sharing that.
                $endgroup$
                – Saeed
                Jan 15 at 17:24










              • $begingroup$
                @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
                $endgroup$
                – Saeed
                Jan 15 at 17:28










              • $begingroup$
                @ Umberto P: Could you help me to prove it?
                $endgroup$
                – Saeed
                Jan 16 at 3:17














              5












              5








              5





              $begingroup$

              In general if $p > 1$ and $u,v ge 0$ then $u^p + v^p le (u+v)^p$. This is evident if $u = 0$, and if $u > 0$ you can divide by $u^p$ to arrive at the equivalent inequality
              $$1 + left( frac vu right)^p le left( 1 + frac vu right)^p.$$



              The function $phi(t) = (1+t)^p - t^p - 1$, $t ge 0$, satisfies $phi'(t) = p(t+1)^{p-1} - pt^{p-1} > 0$ for $t > 0$ because $p-1 > 0$. Thus $phi$ is increasing on $[0,infty)$ so that $phi(t) ge phi(0) = 0$. That is,
              $$1 + t^p le (1+t)^p$$ for all $t ge 0$ and in particular for $t = dfrac vu$.






              share|cite|improve this answer









              $endgroup$



              In general if $p > 1$ and $u,v ge 0$ then $u^p + v^p le (u+v)^p$. This is evident if $u = 0$, and if $u > 0$ you can divide by $u^p$ to arrive at the equivalent inequality
              $$1 + left( frac vu right)^p le left( 1 + frac vu right)^p.$$



              The function $phi(t) = (1+t)^p - t^p - 1$, $t ge 0$, satisfies $phi'(t) = p(t+1)^{p-1} - pt^{p-1} > 0$ for $t > 0$ because $p-1 > 0$. Thus $phi$ is increasing on $[0,infty)$ so that $phi(t) ge phi(0) = 0$. That is,
              $$1 + t^p le (1+t)^p$$ for all $t ge 0$ and in particular for $t = dfrac vu$.







              share|cite|improve this answer












              share|cite|improve this answer



              share|cite|improve this answer










              answered Jan 14 at 21:26









              Umberto P.Umberto P.

              40.1k13368




              40.1k13368












              • $begingroup$
                That's exactly how I was going to do it until I saw that you beat me to it.
                $endgroup$
                – marty cohen
                Jan 14 at 21:27










              • $begingroup$
                I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
                $endgroup$
                – Umberto P.
                Jan 14 at 21:30












              • $begingroup$
                @Umberto P.: Your answer is very elegant. Thank you for sharing that.
                $endgroup$
                – Saeed
                Jan 15 at 17:24










              • $begingroup$
                @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
                $endgroup$
                – Saeed
                Jan 15 at 17:28










              • $begingroup$
                @ Umberto P: Could you help me to prove it?
                $endgroup$
                – Saeed
                Jan 16 at 3:17


















              • $begingroup$
                That's exactly how I was going to do it until I saw that you beat me to it.
                $endgroup$
                – marty cohen
                Jan 14 at 21:27










              • $begingroup$
                I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
                $endgroup$
                – Umberto P.
                Jan 14 at 21:30












              • $begingroup$
                @Umberto P.: Your answer is very elegant. Thank you for sharing that.
                $endgroup$
                – Saeed
                Jan 15 at 17:24










              • $begingroup$
                @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
                $endgroup$
                – Saeed
                Jan 15 at 17:28










              • $begingroup$
                @ Umberto P: Could you help me to prove it?
                $endgroup$
                – Saeed
                Jan 16 at 3:17
















              $begingroup$
              That's exactly how I was going to do it until I saw that you beat me to it.
              $endgroup$
              – marty cohen
              Jan 14 at 21:27




              $begingroup$
              That's exactly how I was going to do it until I saw that you beat me to it.
              $endgroup$
              – marty cohen
              Jan 14 at 21:27












              $begingroup$
              I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
              $endgroup$
              – Umberto P.
              Jan 14 at 21:30






              $begingroup$
              I first saw this argument when I studied $L^p$ space theory back in 19[redacted] and it always stuck with me. It is an elegant application of freshman calculus.
              $endgroup$
              – Umberto P.
              Jan 14 at 21:30














              $begingroup$
              @Umberto P.: Your answer is very elegant. Thank you for sharing that.
              $endgroup$
              – Saeed
              Jan 15 at 17:24




              $begingroup$
              @Umberto P.: Your answer is very elegant. Thank you for sharing that.
              $endgroup$
              – Saeed
              Jan 15 at 17:24












              $begingroup$
              @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
              $endgroup$
              – Saeed
              Jan 15 at 17:28




              $begingroup$
              @Umberto P.: When do we have the reverse inequality $1 + t^p geq (1+t)^p$?
              $endgroup$
              – Saeed
              Jan 15 at 17:28












              $begingroup$
              @ Umberto P: Could you help me to prove it?
              $endgroup$
              – Saeed
              Jan 16 at 3:17




              $begingroup$
              @ Umberto P: Could you help me to prove it?
              $endgroup$
              – Saeed
              Jan 16 at 3:17











              3












              $begingroup$

              It is also a consequence of Minkowski's inequality for $p=frac{3}{2}$. Let
              $$
              stackrel{rightarrow}{a}=(u,0),quadstackrel{rightarrow}{b}=(0,v).
              $$
              Then Minkowski's inequality says
              $$
              left(u^{frac{3}{2}}+v^{{frac{3}{2}}}right)^{frac{2}{3}}=|stackrel{rightarrow}{a}+stackrel{rightarrow}{b}|_{frac{3}{2}}le |stackrel{rightarrow}{a}|_{frac{3}{2}}+|stackrel{rightarrow}{b}|_{frac{3}{2}}=u+v.
              $$






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                I was wondering how we can show it with one famous inequality. Then you did it.
                $endgroup$
                – Saeed
                Jan 14 at 21:43










              • $begingroup$
                Yeah, I had the same mind when I started thinking about it :)
                $endgroup$
                – Song
                Jan 14 at 21:54
















              3












              $begingroup$

              It is also a consequence of Minkowski's inequality for $p=frac{3}{2}$. Let
              $$
              stackrel{rightarrow}{a}=(u,0),quadstackrel{rightarrow}{b}=(0,v).
              $$
              Then Minkowski's inequality says
              $$
              left(u^{frac{3}{2}}+v^{{frac{3}{2}}}right)^{frac{2}{3}}=|stackrel{rightarrow}{a}+stackrel{rightarrow}{b}|_{frac{3}{2}}le |stackrel{rightarrow}{a}|_{frac{3}{2}}+|stackrel{rightarrow}{b}|_{frac{3}{2}}=u+v.
              $$






              share|cite|improve this answer









              $endgroup$













              • $begingroup$
                I was wondering how we can show it with one famous inequality. Then you did it.
                $endgroup$
                – Saeed
                Jan 14 at 21:43










              • $begingroup$
                Yeah, I had the same mind when I started thinking about it :)
                $endgroup$
                – Song
                Jan 14 at 21:54














              3












              3








              3





              $begingroup$

              It is also a consequence of Minkowski's inequality for $p=frac{3}{2}$. Let
              $$
              stackrel{rightarrow}{a}=(u,0),quadstackrel{rightarrow}{b}=(0,v).
              $$
              Then Minkowski's inequality says
              $$
              left(u^{frac{3}{2}}+v^{{frac{3}{2}}}right)^{frac{2}{3}}=|stackrel{rightarrow}{a}+stackrel{rightarrow}{b}|_{frac{3}{2}}le |stackrel{rightarrow}{a}|_{frac{3}{2}}+|stackrel{rightarrow}{b}|_{frac{3}{2}}=u+v.
              $$






              share|cite|improve this answer









              $endgroup$



              It is also a consequence of Minkowski's inequality for $p=frac{3}{2}$. Let
              $$
              stackrel{rightarrow}{a}=(u,0),quadstackrel{rightarrow}{b}=(0,v).
              $$
              Then Minkowski's inequality says
              $$
              left(u^{frac{3}{2}}+v^{{frac{3}{2}}}right)^{frac{2}{3}}=|stackrel{rightarrow}{a}+stackrel{rightarrow}{b}|_{frac{3}{2}}le |stackrel{rightarrow}{a}|_{frac{3}{2}}+|stackrel{rightarrow}{b}|_{frac{3}{2}}=u+v.
              $$







              share|cite|improve this answer












              share|cite|improve this answer



              share|cite|improve this answer










              answered Jan 14 at 21:33









              SongSong

              18.5k21651




              18.5k21651












              • $begingroup$
                I was wondering how we can show it with one famous inequality. Then you did it.
                $endgroup$
                – Saeed
                Jan 14 at 21:43










              • $begingroup$
                Yeah, I had the same mind when I started thinking about it :)
                $endgroup$
                – Song
                Jan 14 at 21:54


















              • $begingroup$
                I was wondering how we can show it with one famous inequality. Then you did it.
                $endgroup$
                – Saeed
                Jan 14 at 21:43










              • $begingroup$
                Yeah, I had the same mind when I started thinking about it :)
                $endgroup$
                – Song
                Jan 14 at 21:54
















              $begingroup$
              I was wondering how we can show it with one famous inequality. Then you did it.
              $endgroup$
              – Saeed
              Jan 14 at 21:43




              $begingroup$
              I was wondering how we can show it with one famous inequality. Then you did it.
              $endgroup$
              – Saeed
              Jan 14 at 21:43












              $begingroup$
              Yeah, I had the same mind when I started thinking about it :)
              $endgroup$
              – Song
              Jan 14 at 21:54




              $begingroup$
              Yeah, I had the same mind when I started thinking about it :)
              $endgroup$
              – Song
              Jan 14 at 21:54











              1












              $begingroup$

              It's
              $$u^3+3u^2v+3uv^2+v^3geq u^3+2sqrt{u^3v^3}+v^3$$ or
              $$3u+3vgeq2sqrt{uv}$$ or
              $$(sqrt{u}-sqrt{v})^2+2u+2vgeq0,$$ which is obvious.






              share|cite|improve this answer









              $endgroup$


















                1












                $begingroup$

                It's
                $$u^3+3u^2v+3uv^2+v^3geq u^3+2sqrt{u^3v^3}+v^3$$ or
                $$3u+3vgeq2sqrt{uv}$$ or
                $$(sqrt{u}-sqrt{v})^2+2u+2vgeq0,$$ which is obvious.






                share|cite|improve this answer









                $endgroup$
















                  1












                  1








                  1





                  $begingroup$

                  It's
                  $$u^3+3u^2v+3uv^2+v^3geq u^3+2sqrt{u^3v^3}+v^3$$ or
                  $$3u+3vgeq2sqrt{uv}$$ or
                  $$(sqrt{u}-sqrt{v})^2+2u+2vgeq0,$$ which is obvious.






                  share|cite|improve this answer









                  $endgroup$



                  It's
                  $$u^3+3u^2v+3uv^2+v^3geq u^3+2sqrt{u^3v^3}+v^3$$ or
                  $$3u+3vgeq2sqrt{uv}$$ or
                  $$(sqrt{u}-sqrt{v})^2+2u+2vgeq0,$$ which is obvious.







                  share|cite|improve this answer












                  share|cite|improve this answer



                  share|cite|improve this answer










                  answered Jan 14 at 21:19









                  Michael RozenbergMichael Rozenberg

                  109k1896201




                  109k1896201























                      1












                      $begingroup$

                      Write $a= sqrt{u}$ and $b= sqrt{v}$, so you have to prove:



                      $$ (a^3+b^3)^2leq (a^2+b^2)^3$$
                      so $$ 2a^3b^3leq 3a^4b^2+3a^2b^4$$



                      so $$2ableq 3a^2+3b^2$$



                      But this is true since $$2ableq a^2+b^2leq 3a^2+3b^2$$






                      share|cite|improve this answer









                      $endgroup$









                      • 1




                        $begingroup$
                        $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                        $endgroup$
                        – marty cohen
                        Jan 14 at 21:26
















                      1












                      $begingroup$

                      Write $a= sqrt{u}$ and $b= sqrt{v}$, so you have to prove:



                      $$ (a^3+b^3)^2leq (a^2+b^2)^3$$
                      so $$ 2a^3b^3leq 3a^4b^2+3a^2b^4$$



                      so $$2ableq 3a^2+3b^2$$



                      But this is true since $$2ableq a^2+b^2leq 3a^2+3b^2$$






                      share|cite|improve this answer









                      $endgroup$









                      • 1




                        $begingroup$
                        $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                        $endgroup$
                        – marty cohen
                        Jan 14 at 21:26














                      1












                      1








                      1





                      $begingroup$

                      Write $a= sqrt{u}$ and $b= sqrt{v}$, so you have to prove:



                      $$ (a^3+b^3)^2leq (a^2+b^2)^3$$
                      so $$ 2a^3b^3leq 3a^4b^2+3a^2b^4$$



                      so $$2ableq 3a^2+3b^2$$



                      But this is true since $$2ableq a^2+b^2leq 3a^2+3b^2$$






                      share|cite|improve this answer









                      $endgroup$



                      Write $a= sqrt{u}$ and $b= sqrt{v}$, so you have to prove:



                      $$ (a^3+b^3)^2leq (a^2+b^2)^3$$
                      so $$ 2a^3b^3leq 3a^4b^2+3a^2b^4$$



                      so $$2ableq 3a^2+3b^2$$



                      But this is true since $$2ableq a^2+b^2leq 3a^2+3b^2$$







                      share|cite|improve this answer












                      share|cite|improve this answer



                      share|cite|improve this answer










                      answered Jan 14 at 21:23









                      Maria MazurMaria Mazur

                      48.5k1260121




                      48.5k1260121








                      • 1




                        $begingroup$
                        $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                        $endgroup$
                        – marty cohen
                        Jan 14 at 21:26














                      • 1




                        $begingroup$
                        $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                        $endgroup$
                        – marty cohen
                        Jan 14 at 21:26








                      1




                      1




                      $begingroup$
                      $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                      $endgroup$
                      – marty cohen
                      Jan 14 at 21:26




                      $begingroup$
                      $2ableq 3a^2+3b^2 iff 0 le (a-b)^2+2a^2+2b^2$.
                      $endgroup$
                      – marty cohen
                      Jan 14 at 21:26











                      0












                      $begingroup$

                      I would raise each side to the $3^{rd}$ power and expand:



                      $$u^3+2(uv)^{frac 32} + v^3 le u^3 + 3u^2v + 3uv^2 + v^3$$



                      Which simplifies to



                      $$2(uv)^{frac 32} le 3uv(u+v)$$



                      Which is



                      $$2sqrt{uv} le 3u+3v$$



                      And if you are familiar with the famous inequality $a^2+b^2 ge 2ab$ (which is gotten from $(a-b)^2 ge 0$) then you see that $2sqrt{uv} le 3u+3v$ as well.






                      share|cite|improve this answer









                      $endgroup$


















                        0












                        $begingroup$

                        I would raise each side to the $3^{rd}$ power and expand:



                        $$u^3+2(uv)^{frac 32} + v^3 le u^3 + 3u^2v + 3uv^2 + v^3$$



                        Which simplifies to



                        $$2(uv)^{frac 32} le 3uv(u+v)$$



                        Which is



                        $$2sqrt{uv} le 3u+3v$$



                        And if you are familiar with the famous inequality $a^2+b^2 ge 2ab$ (which is gotten from $(a-b)^2 ge 0$) then you see that $2sqrt{uv} le 3u+3v$ as well.






                        share|cite|improve this answer









                        $endgroup$
















                          0












                          0








                          0





                          $begingroup$

                          I would raise each side to the $3^{rd}$ power and expand:



                          $$u^3+2(uv)^{frac 32} + v^3 le u^3 + 3u^2v + 3uv^2 + v^3$$



                          Which simplifies to



                          $$2(uv)^{frac 32} le 3uv(u+v)$$



                          Which is



                          $$2sqrt{uv} le 3u+3v$$



                          And if you are familiar with the famous inequality $a^2+b^2 ge 2ab$ (which is gotten from $(a-b)^2 ge 0$) then you see that $2sqrt{uv} le 3u+3v$ as well.






                          share|cite|improve this answer









                          $endgroup$



                          I would raise each side to the $3^{rd}$ power and expand:



                          $$u^3+2(uv)^{frac 32} + v^3 le u^3 + 3u^2v + 3uv^2 + v^3$$



                          Which simplifies to



                          $$2(uv)^{frac 32} le 3uv(u+v)$$



                          Which is



                          $$2sqrt{uv} le 3u+3v$$



                          And if you are familiar with the famous inequality $a^2+b^2 ge 2ab$ (which is gotten from $(a-b)^2 ge 0$) then you see that $2sqrt{uv} le 3u+3v$ as well.







                          share|cite|improve this answer












                          share|cite|improve this answer



                          share|cite|improve this answer










                          answered Jan 14 at 21:23









                          OviOvi

                          12.4k1040114




                          12.4k1040114















                              Popular posts from this blog

                              Human spaceflight

                              Can not write log (Is /dev/pts mounted?) - openpty in Ubuntu-on-Windows?

                              File:DeusFollowingSea.jpg