# Questions d'entretiens - Research staff member

# 199

Questions d'entretien pour Research Staff Member partagées par les candidats## Principales questions d'entretien

### Gaussian linear models are often insufficient in practical applications, where noise can be heavy- tailed. In this problem, we consider a linear model of the form yi = a · xi + b + ei. The (ei) are independent noise from a distribution that depends on x as well as on global parameters; however, the noise distribution has conditional mean zero given x. The goal is to derive a good estimator for the parameters a and b based on a sample of observed (x, y) pairs. 1.1 Instructions: 1. Load the data, which is provided as (x, y) pairs in CSV format. Each file contains a data set generated with different values of a and b. The noise distribution, conditional on x, is the same for all data sets. 2. Formulate a model for the data-generating process. 3. Based on your model, formulate a loss function for all parameters: a, b, and any additional parameters needed for your model. 4. Solve a suitable optimization problem, corresponding to your chosen loss function, to obtain point estimates for the model parameters. 5. Formulate and carry out an assessment of the quality of your parameter estimates. 6. Try additional models if necessary, repeating steps 2 − 5.

3 réponses↳

I think we need to use Generalized Method of Moments to get the estimates. Since E[e|x] = 0, we have E[h(x)e] = 0 by the law of iterated expectation for any give function h(x). Now we need to find a best function h*(x) such that it will give you efficient GMM estimator. Moins

↳

Actually, you will get least squares estimate as the best estimator in the following sense: y = ax+b+e E(e|x)=0 For any h(x), E(h(x)*e) = E(E(h(x)*e)|x) (where the outer expectation is over X E(h(x)*e|x) = h(x)*E(e|x) = 0 Therefore E(h(x)*e)=0 Take h(x) = y-a-b*x The moment condition is: E(e*(y-a-b*x))=0 This would lead to Least Squares. Moins

↳

I believe the true model was y = ax + b + sigma*(x^2). You can use least squares to define the likelihood or use an L1 penalty. Moins

### What are your long-term career objectives? What are your strengths? Weaknesses? What functions or tasks in your last position didn’t you like? Why? Why are you changing positions? What is your most difficult supervisory responsibility? Why? What interests you about this position/this college/this department? Why should I hire you (vs. other candidates)? Describe your research.

1 réponses↳

Embark on independent career as faculty at UCSD. Persistence, accuracy, diplomacy. My weakness is that I sometimes become emotionally invested in a project up to a point that I feel disappointed when it doesn’t turn out well. Cleaning sea urchin tanks after the last person neglected them for a week or more. It is the end of my postdoctoral research career and I am ready to proceed to an independent career as faculty. Managing the “lab family” when there are character clashes. Figuring out what the talents are. It might take a while and sometimes too long to figure it out. Students might get discouraged on their way to find out. We have to walk the thin line of pushing and pulling them to stay on track. I have found that prestigious, successful labs often draw highly skilled, strong willed people. The dynamic between personalities can be a source of contention and it takes diplomacy and mutual understanding to achieve a positive work environment. It is the position in AQUATOX. I feel like this area of science has barely had its surface scratched, and it would be exciting to pioneer further developments in the field. I am drawn to the combination of so many disciplines, including marine sciences, toxicology, and biochemistry. This offers the opportunity to cross-collaborate and integrate interdisciplinary research. Because I have a track record in the competencies you are seeking including a strong background in biochemistry and biophysical analysis techniques. I have done national and international research in Germany and in the US for over 10 years now. I have built up strong ties to faculty and government officials at UCSB, SIO, and through my panel work at CCST, the City of San Diego, and Portland State University. In addition, I have ongoing collaborations with toxicologists and biophysicists in Germany, Rumania, and the Netherlands. My research interests focus on understanding how xenobiotics enter and accumulate in marine organisms. The general questions that I am addressing include: What kind of transport systems are involved in pollutant recognition and exclusion from cells? Why do transporters fail to recognize persistent organic pollutants? Can we determine the structural basis for pollutant/transporter (non-)interactions? Are these interaction patterns conserved across species and thus allow for the prediction of their bioaccumulation? And finally, can we use structural information for the design of safer industrial chemicals? Moins

### Went over my resume and experience, why I wanted to work there, why I thought I was a good fit etc.

1 réponses↳

I was prepared to talk about these things and had no trouble.

### If you find a coworker who is going against procedure, what would you do?

1 réponses↳

I said I would ask them about it to see if the procedure has changed or been updated, and if not explain why the current procedure is in place and why we should follow it. Moins

### Suppose the driver of a push-button ignition vehicle finds his/her vehicle is suddenly accelerating out of control. What would their first reaction be? What improvements need to be made to today's systems to address this?

1 réponses↳

If the driver is outside the vehicle, his/her initial reaction would run after it. If the driver is in the driver's seat, his/her initial reaction would be "slam on foot padle[s]" (hopefully the brakes) and grab the steering wheel. Excluding involuntary/limbic/fight or flight'reactions. Hmm, today's systems I shall first assume refers to the vehicle. In that case, A steering sensor which cuts power to the engine if the wheel has close to zero resistance relative to its torque baseline as the vehicle accelerates. That same system can be applied to engage the brakes. If we are talking about systems in general, the dealership can simply inform/emphasize to the buyer that the vehicle doesn't use keys... Moins

### Questions regarding machine learning and text mining. Conceptual questions about supervise machine learning algorithms

1 réponses↳

How will you handle skewed datasets?

### One unexpected question was: how to measure the length of Huang He River.

1 réponses↳

it's a kind of problem solving question. Thinks of every possible methods to test the length of the river. Moins

### Questions regarding bayesian statistics which I was unfamiliar with.

1 réponses↳

I couldn't answer really but it ended up being more of a discussion and the researchers teaching me which was nice. Moins