Trial and Error - Actually What is Success | Amy Hoover | TEDxNJIT
The speaker argues that relying on single metrics like standardized tests or citation counts to define professional success is flawed because life and complex fields are too nuanced for single measures. Using Professor A (privilege-preordained success) and Professor K (struggle-overcoming-obstacles) as examples, the speaker advises embracing "trial and error" and evaluating success across many different dimensions rather than relying on the first outcome or a single failure. ## Theses & Positions - Quantifying success in professions like teaching (standardized test scores), policing (number of arrests), or research (number of citations/funding) overlooks the inherently complex nature of the problems involved. - Success narratives can be drastically different: one can be "a story of privilege that preordained success," while another is a "struggle of trial and error and the desire to succeed and overcome different obstacles." - Single failures should not define an individual's value because life contains too much randomness. - Success requires evaluating strategies across a variety of dimensions rather than optimizing for a single goal (e.g., best aggressive vs. best control strategy). ## Concepts & Definitions - **Objective:** What people inherently see when describing things with numbers. - **R1 institution:** A designation for a top four research university in the United States. - **Evolutionary computation:** A type of AI used in the speaker's research. - **Reinforcement learning:** A type of AI used in the speaker's research. ## Mechanisms & Processes - **Evaluating decks in Hearthstone:** Testing a card deck against the same opponent multiple times (close to 400) to determine if observed differences in wins/losses are statistically significant. - **Multi-dimensional success evaluation:** Instead of searching for the best outcome in one dimension (e.g., aggressive play), searching across a variety of different measures of success. - **Statistical evaluation:** Determining if a difference in performance is statistically significant after repeated trials. ## Named Entities - **Professor A:** Successful professor from a middle-class family in a sunny Florida town; graduated valedictorian, published first paper as undergrad, received NSF graduate research fellowship, taught at an R1 institution. - **Professor K:** Studied medicine against her interest; father died in a car accident; mother died in a separate car accident; graduated after six years of undergraduate study. ## Numbers & Data - Age when Professor A graduated high school: Implied success path. - Professor K's undergraduate duration: **six years**. - Card game testing requirement: Close to **400** evaluations to test for significance. ## Examples & Cases - **Professor A:** Represents a narrative of privilege leading to success. - **Professor K:** Represents a narrative of struggle, trial and error, and resilience. - **Hearthstone:** A card game used in AI research to test deck strength. - **Famous rejected papers:** Examples of research papers that were initially rejected from conferences/journals but later became famous (to illustrate the failure narrative). ## Tools, Tech & Products - **Hearthstone:** Card game used for AI simulation. - **AI algorithms:** Used in the speaker's research, specifically mentioning evolutionary computation and reinforcement learning. ## Counterarguments & Caveats - The societal tendency to quantify success risks ignoring the difficulty of the underlying problems being studied. - The comparison between Professor A and Professor K, while useful for illustration, is presented as a narrative choice rather than an objective truth about achievement. ## Methodology - The speaker's research methodology involves running simulations within the game Hearthstone using AI to test card decks, requiring evaluation across a large sample size (approaching 400 repetitions) to distinguish noise from genuine performance variation. - The speaker's life experience serves as a method for drawing out actionable advice regarding career planning. ## Conclusions & Recommendations - Do not let a single failure define your value. - Embrace "trial and error" and "continue to explore different avenues of what it means for you to personally be successful." - Approach professional evaluation by considering success over a "large variety of different ways of defining what success means." ## Implications & Consequences - The metrics currently used to evaluate professionals (teachers, police, researchers) may be fundamentally flawed or unfair. - True understanding of a strategy's value requires extensive, varied testing. ## Verbatim Moments - *"there's some sort of objective in that they are objective"* - *"we're ignoring the fact that in the in the first place that that these are hard problems to solve and likely are not able to really be distilled in this way"* - *"this is a story of privilege that preordained success in some ways where this this road to really the road to success has has been paved by privilege in the other is a more of a struggle of trial and error and the desire to succeed and overcome different obstacles in her life to be resilient"* - *"the one failure does not mean that you are a failure it's not scientifically significant and it shouldn't define you"* - *"I need to do in my research is then evaluate these decks a large number of times close to 400 to decide whether or not this difference is significant"* - *"there can be many definitions of success"* - *"it can take many attempts and there are many different ways to evaluate success"*