feat(codeforces): 970

posts/algorithms/competitive-programming-log.html

@@ -24,6 +24,71 @@
           <h1 class="post-title">competitive programming log</h1>
         </header>
         <article class="post-article">
+          <h2>
+            <a href="https://codeforces.com/contest/2008" target="_blank"
+              >970 (div. 3)</a
+            >
+            <span class="post-meta">
+              <time datetime="2025-04-21">04/21/2025</time>
+            </span>
+          </h2>
+          <div>
+            ~1450 performance. OK contest, not very exhilirating. Hedonistic
+            treadmill at work with me solving A-F on Div. 3 when a 2 weeks ago I
+            couldn't even do that on Div. 4. Definitely need to upsolve this to
+            improve my ability at expressing ideas simply.
+          </div>
+          <blockquote>
+            From now on, prioritize actually <i>learning</i> and problem-solving
+            in a pressurized format (we all die eventually, everything is under
+            pressure).
+            <ul>
+              <li>Never report rating in these logs</li>
+              <li>
+                Implement all dsa from scratch (modular arithmetic, data
+                structures, etc.)
+              </li>
+            </ul>
+          </blockquote>
+          <ol>
+            <li>
+              A: somehow got a wrong answer. unbelievable, brute forced it. math
+              is poor. why reason/be ensure when you can brute force?
+            </li>
+            <li>
+              B: didn't outline a simple strategy and got cooked. For example, I
+              didn't check that the grid was a square after redoing my
+              implementation for the third time.
+            </li>
+            <li>C: trivial</li>
+            <li>
+              D: tried a dsu approach before realizing the graph is not
+              direction. DP + DFS + DSU weakness all combined to confuse me. In
+              retrospect, just think about solving the problem in one way. Model
+              it as a graph, do tortoise and hare, or do DP.
+              <b>One approach at a time</b>. I tangibly need to improve my
+              reasoning on harder problems when there are many moving parts.
+              However, I'm unsure of how to move forward here because I'm
+              fighting an uphill battle against a) my basic understanding of
+              algorithms and b) my actual reasoning ability. I need to improve
+              both the basics and my ability to think when facing <i>new</i>
+              problems/frameworks&mdash;that's the sign of a good
+              problem-solver.
+            </li>
+            <li>
+              E: realized the greedy approach + pref/post-fix greedy on deletion
+              but implementation absolutely annahilated me here. Further,
+              <b>my exchange argument/greedy proofs are nightmarishly bad</b>
+              and I resulted in just "trusting" the two-max greedy approach.
+              This also harms my ability/intuition for dp/not doing dp. Work on
+              this.
+            </li>
+            <li>
+              F: easy, counting pairs and modular arithmetic. However, I had to
+              copy the modular arithmetic online. My grasp of this is still not
+              good enough because I haven't practiced enough number theory.
+            </li>
+          </ol>
           <h2>
             <a href="https://codeforces.com/contest/2094" target="_blank"
               >937 (div. 4)</a
@@ -1246,8 +1311,8 @@
               logs, and more. For now, I'll be practicing on
               <a href="https://codeforces.com" target="_blank">CodeForces</a>
               (account
-              <a href="https://codeforces.com/profile/sigill" target="_blank"
+              <a href="https://codeforces.com/profile/sigpipe" target="_blank"
-                >sigill</a
+                >sigpipe</a
               >) and <a href="https://cses.fi" target="_blank">CSES</a>, using
               the
               <a href="https://cses.fi/book/book.pdf" target="_blank"